feat(files): orchestrate connector folder sync

.gitea/workflows/module-package-release.yml

@@ -0,0 +1,270 @@
+name: Module Package Release
+
+on:
+  push:
+    tags:
+      - "v*"
+  workflow_dispatch:
+    inputs:
+      release_tag:
+        description: Existing protected version tag to publish
+        required: true
+        type: string
+
+jobs:
+  publish-packages:
+    runs-on: ubuntu-latest
+    env:
+      GITEA_REPOSITORY: ${{ gitea.repository }}
+    steps:
+      - uses: actions/checkout@34e114876b0b11c390a56381ad16ebd13914f8d5
+        with:
+          fetch-depth: 0
+      - uses: actions/setup-python@a26af69be951a213d495a4c3e4e4022e16d87065
+        with:
+          python-version: "3.12"
+      - uses: actions/setup-node@49933ea5288caeca8642d1e84afbd3f7d6820020
+        with:
+          node-version: "22"
+      - name: Select and validate protected release tag
+        shell: bash
+        env:
+          REQUESTED_TAG: ${{ inputs.release_tag }}
+          TRIGGER_TAG: ${{ gitea.ref_name }}
+        run: |
+          set -euo pipefail
+          tag="${REQUESTED_TAG:-$TRIGGER_TAG}"
+          case "$tag" in
+            v[0-9]*.[0-9]*.[0-9]*) ;;
+            *) echo "Release tag must start with a SemVer-shaped vX.Y.Z value" >&2; exit 1 ;;
+          esac
+          git fetch --force origin "refs/tags/$tag:refs/tags/$tag" refs/heads/main:refs/remotes/origin/main
+          tag_commit="$(git rev-list -n 1 "$tag")"
+          git merge-base --is-ancestor "$tag_commit" refs/remotes/origin/main || {
+            echo "Release tag is not contained in main" >&2
+            exit 1
+          }
+          git checkout --detach "$tag"
+          printf 'RELEASE_TAG=%s\n' "$tag" >> "$GITEA_ENV"
+          printf 'SOURCE_DATE_EPOCH=%s\n' "$(git show -s --format=%ct HEAD)" >> "$GITEA_ENV"
+      - name: Validate package versions
+        run: |
+          python - <<'PY'
+          import json
+          from pathlib import Path
+          import os
+          import re
+          import tomllib
+
+          tag = os.environ["RELEASE_TAG"]
+          expected = tag.removeprefix("v")
+          project = tomllib.loads(Path("pyproject.toml").read_text(encoding="utf-8"))["project"]
+          if project.get("version") != expected:
+              raise SystemExit(f"pyproject version {project.get('version')!r} does not match {tag}")
+          if re.fullmatch(r"govoplan-[a-z0-9-]+", str(project.get("name", ""))) is None:
+              raise SystemExit("Python distribution name must use the govoplan-* namespace")
+          webui = Path("webui/package.json")
+          if webui.is_file():
+              package = json.loads(webui.read_text(encoding="utf-8"))
+              if package.get("version") != expected:
+                  raise SystemExit(f"WebUI version {package.get('version')!r} does not match {tag}")
+              if re.fullmatch(r"@govoplan/[a-z0-9-]+-webui", str(package.get("name", ""))) is None:
+                  raise SystemExit("WebUI package name must use the @govoplan/*-webui namespace")
+              release = Path("webui/package.release.json")
+              if release.is_file():
+                  release_package = json.loads(release.read_text(encoding="utf-8"))
+                  if (
+                      release_package.get("name") != package.get("name")
+                      or release_package.get("version") != expected
+                  ):
+                      raise SystemExit("WebUI release package identity does not match package.json and the release tag")
+          PY
+      - name: Build immutable package artifacts
+        shell: bash
+        run: |
+          set -euo pipefail
+          python -m pip install --disable-pip-version-check build==1.5.0 twine==7.0.0
+          rm -rf dist .package-webui
+          python -m build --wheel --outdir dist
+          python -m twine check dist/*.whl
+          if [[ -f webui/package.json ]]; then
+            mkdir .package-webui
+            cp -a webui/. .package-webui/
+            rm -rf .package-webui/node_modules .package-webui/dist
+            if [[ -f .package-webui/package.release.json ]]; then
+              cp .package-webui/package.release.json .package-webui/package.json
+            fi
+          node <<'NODE'
+          const fs = require("node:fs");
+          const path = ".package-webui/package.json";
+          const packageJson = JSON.parse(fs.readFileSync(path, "utf8"));
+          const groups = ["dependencies", "optionalDependencies", "peerDependencies"];
+          for (const group of groups) {
+            for (const [name, specifier] of Object.entries(packageJson[group] || {})) {
+              if (!name.startsWith("@govoplan/")) continue;
+              if (typeof specifier !== "string") {
+                throw new Error(`${group}.${name} must use a string version`);
+              }
+              const packageSlug = name.slice("@govoplan/".length);
+              if (!packageSlug.endsWith("-webui")) {
+                throw new Error(`${group}.${name} is outside the WebUI package namespace`);
+              }
+              const repository = `govoplan-${packageSlug.slice(0, -"-webui".length)}`;
+              const escapedRepository = repository.replace(/[.*+?^${}()|[\]\\]/g, "\\$&");
+              const gitTag = specifier.match(
+                new RegExp(
+                  `^git\\+(?:ssh://git@|https://)git\\.add-ideas\\.de/(?:GovOPlaN|add-ideas)/${escapedRepository}\\.git#v([0-9]+\\.[0-9]+\\.[0-9]+)$`,
+                ),
+              );
+              if (gitTag) {
+                packageJson[group][name] = gitTag[1];
+                continue;
+              }
+              if (specifier.startsWith("file:") || specifier.startsWith("git+")) {
+                throw new Error(
+                  `${group}.${name} must resolve to an exact registry version for publication`,
+                );
+              }
+            }
+          }
+          delete packageJson.private;
+          fs.writeFileSync(path, `${JSON.stringify(packageJson, null, 2)}\n`);
+          NODE
+            npm pkg delete private --prefix .package-webui
+            (cd .package-webui && npm pack --ignore-scripts --pack-destination ../dist)
+          fi
+          python - <<'PY'
+          import hashlib
+          import json
+          from pathlib import Path
+          import os
+          import subprocess
+
+          artifacts = []
+          for path in sorted(Path("dist").iterdir()):
+              if path.suffix not in {".whl", ".tgz"}:
+                  continue
+              digest = hashlib.sha256(path.read_bytes()).hexdigest()
+              artifacts.append({"filename": path.name, "sha256": digest, "size": path.stat().st_size})
+          payload = {
+              "schema_version": "1",
+              "repository": os.environ["GITEA_REPOSITORY"],
+              "tag": os.environ["RELEASE_TAG"],
+              "commit": subprocess.check_output(["git", "rev-parse", "HEAD"], text=True).strip(),
+              "artifacts": artifacts,
+          }
+          Path("dist/package-artifacts.json").write_text(
+              json.dumps(payload, indent=2, sort_keys=True) + "\n",
+              encoding="utf-8",
+          )
+          PY
+      - name: Retain package hash evidence
+        uses: actions/upload-artifact@a8a3f3ad30e3422c9c7b888a15615d19a852ae32
+        with:
+          name: module-packages-${{ gitea.ref_name }}
+          path: dist/package-artifacts.json
+      - name: Check immutable registry state
+        shell: bash
+        env:
+          PACKAGE_TOKEN: ${{ secrets.GOVOPLAN_PACKAGE_TOKEN }}
+        run: |
+          set -euo pipefail
+          test -n "$PACKAGE_TOKEN"
+          python - <<'PY'
+          import hashlib
+          import json
+          import os
+          from pathlib import Path
+          import tomllib
+          from urllib.error import HTTPError
+          from urllib.parse import quote
+          from urllib.request import Request, urlopen
+
+          api_root = "https://git.add-ideas.de/api/v1/packages/GovOPlaN"
+          token = os.environ["PACKAGE_TOKEN"]
+
+          def should_publish(kind, name, version, path):
+              package_url = "/".join(
+                  (api_root, kind, quote(name, safe=""), quote(version, safe=""), "files")
+              )
+              request = Request(
+                  package_url,
+                  headers={"Accept": "application/json", "Authorization": f"token {token}"},
+              )
+              try:
+                  with urlopen(request, timeout=30) as response:
+                      files = json.load(response)
+              except HTTPError as exc:
+                  if exc.code == 404:
+                      print(f"{kind} package {name}=={version} is not published yet")
+                      return True
+                  raise
+              if not isinstance(files, list) or len(files) != 1:
+                  raise SystemExit(
+                      f"immutable {kind} package {name}=={version} has an unexpected file set"
+                  )
+              expected_sha256 = hashlib.sha256(path.read_bytes()).hexdigest()
+              if files[0].get("sha256") != expected_sha256:
+                  raise SystemExit(
+                      f"immutable {kind} package {name}=={version} already exists with a different SHA-256"
+                  )
+              print(f"verified existing {kind} package {name}=={version} ({expected_sha256})")
+              return False
+
+          project = tomllib.loads(Path("pyproject.toml").read_text(encoding="utf-8"))["project"]
+          wheels = tuple(Path("dist").glob("*.whl"))
+          if len(wheels) != 1:
+              raise SystemExit("release build must contain exactly one wheel")
+          publish_pypi = should_publish(
+              "pypi", str(project["name"]), str(project["version"]), wheels[0]
+          )
+
+          tarballs = tuple(Path("dist").glob("*.tgz"))
+          if len(tarballs) > 1:
+              raise SystemExit("release build must contain at most one npm package")
+          publish_npm = False
+          if tarballs:
+              webui = json.loads(
+                  Path(".package-webui/package.json").read_text(encoding="utf-8")
+              )
+              publish_npm = should_publish(
+                  "npm", str(webui["name"]), str(webui["version"]), tarballs[0]
+              )
+
+          with Path(os.environ["GITEA_ENV"]).open("a", encoding="utf-8") as env_file:
+              env_file.write(f"PUBLISH_PYPI={int(publish_pypi)}\n")
+              env_file.write(f"PUBLISH_NPM={int(publish_npm)}\n")
+          PY
+      - name: Publish wheel and WebUI package
+        shell: bash
+        env:
+          PACKAGE_USERNAME: ${{ secrets.GOVOPLAN_PACKAGE_USERNAME }}
+          PACKAGE_TOKEN: ${{ secrets.GOVOPLAN_PACKAGE_TOKEN }}
+        run: |
+          set -euo pipefail
+          test -n "$PACKAGE_USERNAME"
+          test -n "$PACKAGE_TOKEN"
+          if [[ "$PUBLISH_PYPI" == 1 ]]; then
+            TWINE_USERNAME="$PACKAGE_USERNAME" TWINE_PASSWORD="$PACKAGE_TOKEN" \
+              python -m twine upload --non-interactive \
+                --repository-url https://git.add-ideas.de/api/packages/GovOPlaN/pypi \
+                dist/*.whl
+          else
+            echo "Exact wheel is already present; skipping immutable retry."
+          fi
+          shopt -s nullglob
+          webui_packages=(dist/*.tgz)
+          if (( ${#webui_packages[@]} )) && [[ "$PUBLISH_NPM" == 1 ]]; then
+            npmrc="$(mktemp)"
+            trap 'rm -f "$npmrc"' EXIT
+            chmod 600 "$npmrc"
+            printf '%s\n' \
+              '@govoplan:registry=https://git.add-ideas.de/api/packages/GovOPlaN/npm/' \
+              "//git.add-ideas.de/api/packages/GovOPlaN/npm/:_authToken=$PACKAGE_TOKEN" \
+              > "$npmrc"
+            NPM_CONFIG_USERCONFIG="$npmrc" npm publish "./${webui_packages[0]}" \
+              --ignore-scripts --access public \
+              --registry https://git.add-ideas.de/api/packages/GovOPlaN/npm/
+          elif (( ${#webui_packages[@]} )); then
+            echo "Exact WebUI package is already present; skipping immutable retry."
+          fi

AGENTS.md

@@ -1,5 +1,11 @@
 # GovOPlaN Files Codex Guide
 
+## Documentation Contract
+
+- Treat documentation as part of every behavior change. Update this module's manifest-driven `DocumentationTopic` contributions for affected user and administrator behavior.
+- Keep feature content here; `govoplan-docs` projects it without importing Files internals.
+- Maintain a static user/admin baseline and run `/mnt/DATA/git/govoplan/tools/checks/check-manifest-shapes.py` after behavior or manifest changes.
+
 ## Scope
 
 This repository owns the `files` module: managed file storage APIs, file metadata, shares, uploads/downloads, folder and pattern helpers, backend module manifest, and `@govoplan/files-webui`.

README.md

@@ -55,6 +55,15 @@ for example for managed attachment selection and campaign file sharing. Files
 does not import campaign internals; campaign share/existence checks use the core
 `campaigns.access` capability registered by the campaign module.
 
+Files also publishes the optional `privacy.dsar.files` capability. The Core
+data-subject workflow can use it to collect bounded, tenant-scoped file,
+version, share, folder, evidence, and non-secret configuration metadata for a
+direct membership subject. The provider may revoke a subject-targeted share or
+detach mutable actor references idempotently, but it never exports credential
+material or raw bytes and never bypasses Files retention, legal hold, evidence,
+purge approval, audit, or recovery controls. See the handbook's data-subject
+request coverage section for the review and erasure boundary.
+
 Platform RBAC and governance rules are documented in `govoplan-core/docs/`.
 
 Managed files can carry source provenance for connector and import workflows.
@@ -105,7 +114,8 @@ run profile policy with the `browse` operation, and support Seafile,
 WebDAV/Nextcloud, and SMB when the optional `smb` extra is installed. Seafile
 profiles browse libraries and directories via Seafile's read-only API; profiles
 can still opt into WebDAV browsing by setting `metadata.webdav_endpoint_url` or
-`metadata.browse_protocol` to `webdav`.
+`metadata.browse_protocol` to `webdav`. S3-compatible profiles support bucket
+and prefix browsing when the optional `s3` extra is installed.
 `POST /api/v1/files/connectors/profiles/{profile_id}/import` imports a Seafile
 or WebDAV/Nextcloud/SMB file into managed storage through the same governance,
 conflict handling, source provenance, and connector audit path as direct
@@ -113,17 +123,38 @@ uploads. The Seafile provider uses account-token auth and the native file
 download-link API; Nextcloud and generic WebDAV profiles use authenticated `GET`
 requests against the configured WebDAV endpoint. SMB profiles use
 `smb://server[:port]/share[/path]` endpoints and deployment-owned or encrypted
-stored credentials through `smbprotocol`. Because that SDK cannot accept a
+stored credentials through `smbprotocol`. Files installs a pinned transport for
-preconnected socket and may follow DFS referrals to additional hosts, live SMB
+every initial session, reconnect, alias, and DFS referral target. The deployment
-access fails closed in both public-only and private-network deployments. It will
+private-network policy is re-evaluated immediately before each socket opens.
-remain disabled until every initial connection and referral target can be
-policy-validated and pinned.
 
-The S3 browse/import implementation and S3 managed-storage backend currently
+S3 connector browse/import binds botocore HTTP and HTTPS pools to the same pinned
-fail closed before creating a `boto3` client. Botocore does not yet use the
+socket policy. Retries, redirects, endpoint discovery, bucket aliases, and new
-GovOPlaN pinned HTTP transport and may manage redirects itself, so neither an
+connections are therefore revalidated while TLS keeps the configured hostname
-explicit endpoint nor SDK endpoint discovery is allowed until both peer pinning
+for SNI and certificate checks. Connector clients do not use outbound proxies or
-and redirect revalidation are enforced.
+ambient AWS credential discovery; configure credentials on the governed profile,
+or explicitly use an anonymous profile for public objects. Keep connector spaces
+read-only unless a remote write is explicitly needed.
+An S3 profile with the `write` capability can back a two-way connector space.
+`POST /api/v1/files/connector-spaces/{space_id}/write-back` then conditionally
+creates or replaces one remote object from a managed file, records durable
+recovery intent before the provider effect, and verifies request/content markers
+afterward. Automatic remote delete, rename, move, and ACL propagation remain
+disabled; other connector providers remain read-only.
+An incompatible SDK upgrade fails closed before a usable client/session is
+returned.
+
+Durable platform storage has a separate deployment boundary: installer-owned
+Garage is accepted only at its exact generated endpoint, while an
+operator-selected external backend requires a clean HTTPS origin and
+`FILE_STORAGE_S3_ENDPOINT_TRUSTED=true`. That flag is deployment configuration,
+cannot be supplied through a Files connector profile, and does not replace
+operator responsibility for DNS, certificates, egress, bucket policy,
+versioning, and recovery.
+
+The actual local/S3 backend implementation is owned by Core so Files, Campaign,
+and workers resolve the same object namespace. Files owns file metadata and key
+layout. Node-local storage is supported only for `local` or one-host
+`host-shared` profiles; multi-host `shared` deployments require S3.
 
 Destructive Files-module retirement applies the same credential lifecycle before
 dropping tables. Every remaining Files-owned encrypted connector secret is
@@ -139,17 +170,46 @@ Connector and collaboration ownership boundaries are documented in
 `docs/CONNECTOR_BOUNDARY.md` and `docs/DOCUMENT_COLLABORATION_BOUNDARY.md`.
 The role-adaptive user, administration, integration, and operator guide is the
 [Files handbook](docs/FILES_HANDBOOK.md).
+The Files route, connector surfaces, consequence classes, and pattern-language
+verification are recorded in
+[Files interface pattern migration](docs/INTERFACE_PATTERN_MIGRATION.md).
 
-ZIP uploads are processed without buffering the whole archive in memory. The API
+Archive imports use a two-phase preview and confirmation flow for ZIP, TAR,
-spools incoming ZIP request bodies to a bounded temporary file, then extracts
+TAR.GZ, TAR.BZ2, and TAR.XZ. Requests are spooled to bounded temporary files;
-members with per-file and total extracted-size limits before storing managed
+the server validates paths, entry count, expanded size, and expansion ratio,
-files.
+then returns a 30-minute tenant/user-bound preview token. Confirmation reuploads
+the original archive and stores only the selected members. Password-protected
+ZIP passwords remain request-only and are never included in the preview token.
+
+Managed blob writes and applied orphan cleanup use Core's durable recovery
+ledger. On PostgreSQL, intent, request digests, recovery mode, and a distributed
+lease are committed before physical storage effects; the Files session commit
+verifies database and streamed object evidence, while rollback compensates only
+a newly reserved unreferenced key. Development SQLite records blob intent in
+the caller transaction to avoid its second-writer deadlock, then verifies on
+commit or reconstructs compensation evidence after handled rollback. Because a
+hard loss before that commit can leave an unrecorded object, SQLite requires a
+complete integrity scan after a crash and is not a production recovery profile.
+New object keys are opaque and do not retain the uploaded filename. Uncertain
+or mismatched effects remain visible through Ops.
+
+Operators with `files:file:admin` can run bounded, resumable integrity scans in
+Administration. Scan batches and finding actions carry monotonic revisions;
+stale resume, recheck, or cleanup requests fail before touching object storage.
+Orphan cleanup always requires a dry-run preview followed by separate
+confirmation and records recovery-ledger evidence.
 
 Bulk rename and transfer APIs are owner-scoped: callers must provide the active
 user or group file space with `owner_type` and `owner_id`. The storage layer
 keeps a named legacy file-only helper for historical callers that lack owner
 context, and regression tests cover its write-access checks.
 
+Optional producer modules can persist generated output through the provider-
+neutral `files.artifact_store` capability. Files remains responsible for upload
+authorization, ownership, path normalization, versioning, and blob storage;
+producers receive stable file/version references without importing Files
+internals. See [Generated Artifact Store](docs/GENERATED_ARTIFACT_STORE.md).
+
 ## Release packaging
 
 The repository root includes a `package.json` for git-based WebUI installs. It exports the package `@govoplan/files-webui` from `webui/src` so release builds can depend on tagged git refs instead of local `file:` paths.

docs/CONNECTOR_BOUNDARY.md

@@ -19,6 +19,7 @@ normal product workflows and can share the same governance model:
 - Nextcloud through WebDAV
 - generic WebDAV
 - SMB through the optional `smb` extra
+- S3-compatible stores through the optional `s3` extra
 
 These providers are surfaced through connector descriptors at
 `GET /api/v1/files/connectors/providers`. Provider descriptors declare whether
@@ -60,11 +61,14 @@ Every provider must:
 - use a transport that pins every connection to a policy-validated DNS/IP answer
   and revalidates redirects; SDK transports without that guarantee fail closed
 
-Live SMB access is disabled in all modes until `smbprotocol` initial connections
+SMB initial connections, reconnects, aliases, and DFS referral targets are
-and DFS referral targets can be pinned and policy-validated. An explicit IP is
+created through a Files-owned pinned `smbprotocol` transport. S3 HTTP/HTTPS pools
-not sufficient because the server may still issue a referral to another peer.
+use the equivalent botocore adapter for every connection selected by retries,
-Live S3 access is likewise disabled until `boto3`/botocore can be bound to the
+redirects, endpoint discovery, and virtual-host addressing. Both adapters apply
-pinned transport, including SDK-managed redirects and endpoint discovery.
+the deployment-wide private-network policy at socket creation. S3 keeps the
+configured hostname for TLS SNI and certificate verification, but does not use
+outbound proxies or ambient AWS credential discovery. An incompatible optional
+SDK release fails closed before a usable session or client is returned.
 
 ## Non-Goals For Files
 

docs/CONNECTOR_SPACES.md

@@ -61,11 +61,10 @@ Linked connector space:
 - read-only flag from provider/policy
 - active/deleted state
 
-The linked space should be addressable as a normal file space in the files UI.
+The linked space is addressable as a normal file space in the Files UI. It can
-For the first implementation slice, actions may use the existing connector
+browse the remote folder, synchronize one selected file, or run a bounded manual
-browse/sync APIs and managed file storage. A linked space can therefore browse
+folder synchronization into managed storage. Background sync jobs remain a
-the remote folder and sync selected files into managed storage. Later slices can
+separate scheduling concern.
-add a native remote listing view or background sync jobs.
 
 ## Policy Semantics
 
@@ -105,6 +104,9 @@ Implemented:
 - Files UI sync dialog for choosing a profile, browsing a remote folder, and
   syncing a selected file into a managed destination folder
 - Files UI connector-space view with provider/read-only/manual-sync state
+- bounded folder-level manual sync with recursive traversal, pagination,
+  per-item review outcomes, source-provenance matching, and one summary audit
+  event
 - Docker dev stack smoke checks for WebDAV, Nextcloud, and SMB
 
 Missing:
@@ -137,9 +139,14 @@ Missing:
    - Remaining: edit/manage actions for existing linked spaces.
 
 4. Add sync orchestration.
-   - Manual sync selected file is already available.
+   - Manual selected-file and folder-level synchronization are available.
-   - Add folder-level manual sync.
+   - Folder sync preserves remote relative paths, bounds file count and depth,
-   - Add optional scheduled/background sync with conflict reporting.
+     and reports created, updated, unchanged, skipped, conflict, policy-denied,
+     and failed items without rolling back successful siblings.
+   - The default conflict strategy is `skip`, so unrelated managed files are
+     left untouched. `rename`, `reject`, and `overwrite` remain explicit choices.
+   - Add optional scheduled/background sync separately; it must reuse the same
+     policy, provenance, result, and audit contracts.
 
 5. Add provider-specific expansion.
    - OAuth/secret-store credentials.

docs/FILES_HANDBOOK.md

@@ -1,6 +1,6 @@
 # GovOPlaN Files Handbook
 
-This handbook describes the Files module as implemented in version `0.1.9`.
+This handbook describes the Files module as implemented in version `0.1.18`.
 It is the operational source of truth for users, process owners, administrators,
 operators, auditors, and module integrators. Statements about future behavior
 are marked **planned**; an unmarked statement describes the current code.
@@ -8,7 +8,8 @@ are marked **planned**; an unmarked statement describes the current code.
 Files is a governed snapshot store. It owns managed file content, versions,
 logical folders, shares, source provenance, and the evidence that another
 GovOPlaN module used a particular file version. It can browse selected external
-stores read-only and import a frozen copy. It is not a general remote filesystem,
+stores and import a frozen copy. Explicit, conditional S3 write-back is available
+only through an administrator-enabled two-way connector space. Files is not a general remote filesystem,
 a document collaboration engine, or a records-management system.
 
 ## Choose a reading path
@@ -24,32 +25,65 @@ a document collaboration engine, or a records-management system.
 | Verify a release or scenario | [Acceptance scenarios](#acceptance-scenarios) |
 | Check whether an idea exists today | [Implemented and planned boundary](#implemented-and-planned-boundary) |
 
+The Files-owned interface archetypes, consequence classes, disabled-state
+wording, and verification evidence are recorded in
+[Files Interface Pattern Migration](INTERFACE_PATTERN_MIGRATION.md).
+
 ## The service contract
 
 A managed file is a tenant-scoped logical asset with exactly one user or group
 owner, a normalized path, and a current version. The current version points to a
 blob record containing the storage location, SHA-256 checksum, byte size, and
-content type. Current service paths append a version when connector sync finds
+content type. A protected blob also records its Encryption envelope, protection
+discriminator, stored-ciphertext checksum, and stored-ciphertext size. Current
+service paths append a version when connector sync finds
 changed content; they do not mutate the previous version record.
 
-Content with the same tenant, SHA-256 checksum, and size can reuse one blob.
+Unprotected content with the same tenant, plaintext SHA-256 checksum, size, and
-This is storage deduplication and integrity evidence, not proof of authorship or
+protection discriminator can reuse one blob. Protected content is deduplicated
-source authenticity.
+only inside the same vault/profile discriminator; ciphertext is never silently
+reused across protection boundaries. Plaintext checksums remain semantic
+version evidence, while download and integrity scans verify stored ciphertext
+before asking Encryption to open it. Neither digest proves authorship or source
+authenticity.
 
 The main domain objects are:
 
 | Object | Meaning | Lifecycle today |
 | --- | --- | --- |
-| File asset | The user-facing file identity, owner, logical path, description, metadata, and current version | Created, organized, shared, and soft-deleted |
+| File asset | The user-facing file identity, owner, logical path, description, metadata, current version, retention deadline, legal hold, and lifecycle revision | Created, organized, shared, soft-deleted, restored, or governed-purged |
-| File version | A numbered snapshot of one asset and its blob | Appended by connector sync when bytes change; retained |
+| File version | A numbered snapshot of one asset and its blob | Appended by connector sync when bytes change; retained until its asset is eligible for purge |
-| File blob | Stored bytes plus checksum, size, backend, key, reference count, and optional retention timestamp | Reused within a tenant; no automated garbage collection |
+| File blob | Stored plaintext semantics plus stored-byte integrity, backend key, optional Encryption envelope, reference count, and retention timestamp | Reused only within a tenant and matching protection boundary; collected only after a fenced reference recheck |
-| Folder | An explicit logical path in a user or group space | Created, moved/renamed through organize operations, and soft-deleted |
+| Folder | An explicit logical path in a user or group space | Created, moved/renamed, soft-deleted, and restored when its original path is free |
 | Share | A grant from an asset to a user, group, tenant, or campaign with `read`, `write`, or `manage` permission | Created or updated; no public revocation endpoint yet |
 | Connector profile | A governed external endpoint, scope, optional credential link, policy, and descriptive capabilities | Created, updated, disabled, or credential-scrubbed on deletion |
 | Connector credential | Reusable authentication material, optionally limited to a provider and scope | Encrypted when database-managed; immediately scrubbed on deletion |
 | Connector policy | Allow and deny rules inherited from system through tenant to one leaf scope | Evaluated before configuration and connector use |
-| Connector space | A read-only, manually synchronized remote folder/library linked to a user or group space | Created, updated, disabled, and soft-deleted |
+| Connector space | A manually synchronized remote folder/library linked to a user or group space; read-only by default, with explicit S3 two-way mode available | Created, updated, disabled, soft-deleted, and restored |
 | Campaign attachment use | Evidence connecting a campaign job or entry to an exact asset, version, blob, checksum, and stage | Retained for campaign execution evidence |
+| Form evidence upload grant | A one-use, hash-only bearer grant tied to an exact Form instance/revision, purpose, custodian, size, and media-type policy | Issued for at most 15 minutes, consumed by one managed upload, then retained as evidence provenance |
+
+## Deployment configuration packages
+
+Files registers the `files.configuration` capability for `managed_storage`
+fragments. Managed content storage is deployment-owned: the installer selects
+local persistent storage, managed Garage, or an external S3-compatible service
+and mounts a validated non-secret infrastructure capability receipt. Files
+does not copy that endpoint or its credentials into module-owned tables.
+
+Preflight compares the receipt's `files.storage` capability with the effective
+runtime backend. It validates backend kind, sanitized S3 endpoint, bucket,
+Garage management marker or external trust marker, an absolute persistent path
+for local storage, and the presence of Files-owned `env:` secret references.
+Secret values are never read into a plan, diagnostic, export, or fragment.
+
+When runtime and receipt agree, the plan reports `skip`: the desired binding is
+already effective, and repeated apply is a no-op. A mismatch blocks import and
+explains which deployment setting must be reconciled. The provider deliberately
+does not mutate process environment, migrate stored objects, probe remote
+storage, or reinterpret an infrastructure replacement as safe. Use the Files
+integrity and Ops checks after deployment and complete migration/recovery review
+before changing an active backend.
 
 ## User tasks
 
@@ -72,9 +106,15 @@ Files administrator; governed API operations can administer other tenant-owned
 Files resources when their owner is specified. Linked connector spaces appear
 beside managed spaces when they are active and visible to the user.
 
-A connector space is intentionally read-only. It is a view of an approved
+A connector space is read-only by default. It is a view of an approved remote
-remote location and a starting point for importing or synchronizing selected
+location and a starting point for importing a selected file or synchronizing a
-files into managed storage; it is not a mounted write-through filesystem.
+selected file or bounded folder tree into managed storage. Folder sync preserves
+relative remote paths in the linked owner's managed space and returns a
+reviewable result for every discovered or deliberately skipped item. An
+administrator may opt an S3 space into two-way mode
+only after enabling the profile's `write` capability. Even then, Files exposes
+explicit, conditional file write-back—not a mounted filesystem. Automatic
+remote delete, rename, move, and ACL propagation stay disabled.
 
 ### Upload files
 
@@ -94,21 +134,29 @@ An ordinary upload does not append a new version to an existing asset. The
 `overwrite` strategy retires the old asset at that path. Version-preserving
 updates currently belong to connector sync.
 
-### Upload and unpack ZIP files
+### Preview and unpack archives
 
-The UI can unpack a ZIP upload. The request body is spooled to a bounded
+The UI previews ZIP, TAR, TAR.GZ, TAR.BZ2, and TAR.XZ before writing managed
-temporary file rather than buffered wholly in memory. The defaults are:
+files. Users may select individual files or complete folders. The request is
+spooled to a bounded temporary file rather than buffered wholly in memory. The
+defaults are:
 
-- 250 MiB compressed request and total extracted data;
+- 250 MiB compressed request;
 - 50 MiB per extracted member;
-- 1,000 non-directory members;
+- 2 GiB total expanded data;
-- encrypted archives are rejected;
+- 10,000 declared entries;
-- member paths are normalized and `..` traversal is rejected;
+- 100:1 maximum expansion ratio;
-- the actual bytes read are counted, not only the ZIP header declarations.
+- a 30-minute preview token bound to the tenant, user, archive digest, and
+  destination;
+- password-protected ZIP support with request-only password handling;
+- traversal, duplicate paths, links, devices, and other special entries are
+  rejected;
+- actual bytes read are counted, not only archive header declarations.
 
-The byte limits use `FILE_UPLOAD_ZIP_MAX_BYTES` and
+The browser retains the selected archive and password until confirmation.
-`FILE_UPLOAD_MAX_BYTES`. The 1,000-member limit is currently fixed in the
+Confirmation reuploads the archive, verifies the token and digest, repeats all
-extraction service.
+safety checks, and commits only the selected files. No preview archive or
+password is retained server-side.
 
 ### Organize files and folders
 
@@ -129,6 +177,73 @@ Recursive folder deletion is the default. It soft-deletes the selected folder,
 its child folders, and files below it. A non-recursive delete fails when the
 folder is not empty.
 
+### Restore, retain, hold, and purge
+
+Soft deletion is reversible for callers with `files:file:restore`. Restoring a
+file preserves its asset identity, complete version history, blob references,
+and source provenance. Recursive folder restore reactivates the folder tree and
+its deleted files. Connector-space restore reactivates only the local link.
+Each restore fails if an active resource already occupies the original path or
+label.
+
+Retention and legal hold are independent lifecycle controls. A caller with
+`files:file:retention` supplies the current lifecycle revision, a reason, an
+optional retained-until time, and legal-hold state. Stale revisions fail rather
+than overwriting a concurrent decision.
+
+Hard purge is deliberately separate from ordinary delete:
+
+1. A caller with `files:file:purge` previews 1–100 soft-deleted assets.
+2. The preview reports retention, legal-hold, active-share, Campaign-evidence,
+   and Form-evidence blockers and returns a SHA-256 over current lifecycle and
+   blob-reference state.
+3. Execution requires that exact hash, a stable idempotency key, an approval
+   reference, and the literal `PURGE` confirmation.
+4. The recovery-ledger operation and tenant purge fence are durable before the
+   irreversible database transaction. A stale preview or blocker rejects the
+   operation without erasure.
+5. Purge removes eligible asset, version, and inactive-share rows and
+   recalculates blob reference counts. It does not delete bytes inline.
+6. A separately authorized bounded blob-GC call takes the same per-blob lease
+   used by uploads, rechecks all `FileVersion` references, deletes the exact
+   object, verifies absence, and only then deletes `FileBlob` metadata.
+
+Automatic time-based purge scheduling is not implemented. Operators initiate
+preview, execute, and garbage collection under their local retention process.
+
+### Data-subject request coverage
+
+Files registers `privacy.dsar.files` when the module is active. The provider
+requires a corroborated tenant membership identifier (or a namespaced Files
+user reference), searches only that tenant, and fails explicitly if its bounded
+result limit would be exceeded. It reports managed assets, exact versions,
+folders, user-targeted shares, Form and Campaign evidence, connector
+configuration actor references, and integrity-operation evidence.
+
+The export includes only governed metadata. It never embeds file bytes, blob
+storage keys, passwords, tokens, environment-variable names, secret-provider
+references, or encrypted connector values. A reviewer follows the authorized
+version download route when the file itself must be inspected.
+
+The erasure plan deliberately separates four outcomes:
+
+- an active share aimed at the subject can be revoked idempotently;
+- mutable creator/updater references can be detached after tenant, subject, and
+  current-value revalidation;
+- legal hold, active retention, submitted Form evidence, Campaign delivery
+  evidence, connector configuration history, and integrity evidence are
+  retained with a reason; and
+- unstructured file content, ownership, filenames, and paths require manual
+  review.
+
+DSAR execution never invokes physical byte deletion. If the privacy decision
+authorizes erasure, the operator must use the separate Files soft-delete, purge
+preview, approval, execution, and blob-GC sequence. This preserves its distinct
+authority, evidence blockers, audit trail, distributed fencing, and recovery
+ledger semantics. Email, account, or identity selectors alone are insufficient
+because Files does not import the Access directory; the Access search supplies
+the corroborated membership reference for Files coverage.
+
 ### Find and download files
 
 Files can list by owner and path, use cursor pagination, and consume incremental
@@ -169,7 +284,35 @@ inside the chosen owner space:
 - identical checksum and size updates provenance and returns `unchanged`;
 - changed bytes append a version and return `updated`.
 
-Browse, import, and sync never write, rename, or delete the remote source.
+Folder-level manual sync uses
+`POST /api/v1/files/connector-spaces/{space_id}/sync`. The interactive request
+is bounded by file count and traversal depth, follows provider pagination, and
+can include or exclude subfolders. Every file is isolated in a database
+savepoint, so one conflict, policy denial, invalid provider result, or transport
+failure remains reviewable without discarding successful siblings. Results are
+classified as `created`, `updated`, `unchanged`, `skipped`, `conflict`,
+`policy_denied`, or `failed`. The UI defaults unrelated target collisions to
+`skip`; `rename`, `reject`, and `overwrite` require an explicit operator choice.
+The response marks a run as truncated when its bound is reached so the operator
+can continue with a narrower remote folder. This is interactive orchestration,
+not background scheduling.
+
+Browse, import, and inbound sync never mutate the remote source. An S3 space in
+explicit two-way mode can write a selected managed file to one remote object
+path through `POST /api/v1/files/connector-spaces/{space_id}/write-back`.
+Creating a path uses `If-None-Match`; overwriting requires the currently
+observed ETag or version and uses a conditional request. The provider object is
+then re-read and must contain both the expected content digest and the recovery
+operation marker. Uncertain or mismatching outcomes remain fenced and visible
+in Ops. Remote deletion, rename, move, and permission propagation are disabled.
+
+Connector administration separates endpoint profiles, reusable credentials,
+and inherited policy. Ordinary setup uses typed fields and provider discovery;
+provider metadata JSON is available only under advanced compatibility options.
+Read-only deployment entries explain where they must be changed, and disabled
+actions identify the missing permission, target, input, or running operation.
+The contextual help icon opens the configured Help Center topic when Docs is
+enabled and the hosted GovOPlaN documentation otherwise.
 
 ## Process perspective
 
@@ -187,9 +330,10 @@ The managed-file flow is:
 8. Connector-originated operations also emit their connector audit evidence.
 
 Blob storage is not part of the database transaction. An object may therefore
-be left without committed metadata after a process or database failure. No
+be left without committed metadata after a process or database failure. The
-automatic orphan reconciliation exists yet; operators must account for this in
+operator integrity API scans database blobs and the tenant storage prefix in
-integrity checks and retention plans.
+bounded, resumable phases. It reports orphan objects before any cleanup and
+never deletes them as part of a scan.
 
 ### Governed connector import
 
@@ -266,10 +410,14 @@ those relationships first.
 | --- | --- |
 | `files:file:read` | List and inspect accessible files, folders, spaces, and visible connectors |
 | `files:file:download` | Download an accessible current version or ZIP archive |
-| `files:file:upload` | Upload managed assets and import/sync selected connector files |
+| `files:file:upload` | Upload managed assets and import/sync selected connector files or bounded connector folders |
 | `files:file:organize` | Create folders, rename, move/copy, and manage linked connector spaces |
 | `files:file:share` | Create or update file shares |
 | `files:file:delete` | Soft-delete accessible writable files and folders |
+| `files:file:restore` | Restore owned or administered soft-deleted files, folders, and connector-space links |
+| `files:file:retention` | Set retention deadlines and legal holds with optimistic revision checks |
+| `files:file:purge` | Preview and execute irreversible metadata purge and collect unreferenced blobs |
+| `files:connector:write` | Write an accessible managed file to an explicitly writable connector space |
 | `files:file:admin` | Administer all Files spaces and connector settings in the active tenant |
 
 The `file_manager` role template grants all normal file operations except
@@ -289,12 +437,14 @@ Keep these definitions separate:
   local policy, and descriptive operation capabilities;
 - a **policy** restricts what a scope may configure or use;
 - a **connector space** links one approved profile/library/path to one user or
-  group and always uses manual, read-only synchronization today.
+  group, uses manual synchronization, and is read-only unless explicitly
+  configured for the supported two-way S3 write boundary.
 
-Profile capability values such as `browse`, `import`, and `sync` are stored and
+Profile capability values such as `browse`, `import`, `sync`, and `write` are
-returned, but they are descriptive today. Provider implementation and policy
+stored and returned. `write` is additionally enforced for S3 write-back, but
-checks enforce actual availability; do not use the capability list as the sole
+provider implementation, connector-space mode, operation permission, resource
-security control.
+access, and inherited policy all remain mandatory; never use the capability
+list as the sole security control.
 
 Profiles, credentials, and policies support `system`, `tenant`, `user`,
 `group`, and `campaign` scopes. A normal user sees system and active-tenant
@@ -378,15 +528,15 @@ in a development/test runtime.
 | Seafile | Read-only native API browse/download-link import and manual sync implemented using the pinned HTTP transport; WebDAV opt-in supported |
 | Nextcloud | Read-only WebDAV browse/import/manual sync implemented using the pinned HTTP transport |
 | Generic WebDAV | Read-only browse/import/manual sync implemented using the pinned HTTP transport |
-| SMB | Browse/import code and descriptor implemented, but every live connection fails closed until initial connections and DFS referrals can be policy-validated and pinned |
+| SMB | Read-only browse/import/manual sync implemented through a pinned smbprotocol transport for initial peers, reconnects, aliases, and DFS referral targets |
-| S3 connector | Browse/import code and descriptor implemented, but every live SDK connection fails closed until botocore connections, redirects, and endpoint discovery can be validated and pinned |
+| S3 connector | Bucket/prefix browse, import, manual inbound sync, and explicit conditional write-back implemented through pinned botocore pools covering retries, redirects, endpoint discovery, and provider aliases; automatic remote delete/rename/move/ACL propagation disabled |
 | SharePoint and OneDrive | Provider keys/descriptors reserved; live Microsoft Graph browse/import is planned |
 | NFS and local connector | Described as optional future providers; the local managed-storage backend is a different feature |
 
 Provider descriptors are available from
 `GET /api/v1/files/connectors/providers`. Use their `implemented`, `installed`,
-and support fields for display, but expect a fail-closed transport error where
+and support fields for display. An incompatible optional SDK release fails closed
-the table above says live access is disabled.
+before it can return a usable client or session.
 
 ## Operator runbook
 
@@ -398,8 +548,15 @@ the table above says live access is disabled.
 | `FILE_STORAGE_LOCAL_ROOT` | `runtime/files` | Primary local write/read root |
 | `FILE_STORAGE_LOCAL_FALLBACK_ROOTS` | empty | Comma-separated older read-only roots checked after the primary root |
 | `FILE_STORAGE_S3_ENDPOINT_URL` and related `FILE_STORAGE_S3_*` values | deployment-specific | S3-compatible endpoint, region, credentials, and bucket |
-| `FILE_UPLOAD_MAX_BYTES` | 50 MiB | Direct-upload and ZIP-member maximum |
+| `FILE_STORAGE_S3_DEPLOYMENT_MANAGED` | `false` | Installer-only trust marker for the exact `http://garage:3900` service; never use it for another endpoint |
-| `FILE_UPLOAD_ZIP_MAX_BYTES` | 250 MiB | ZIP request and extracted-total maximum |
+| `FILE_STORAGE_S3_ENDPOINT_TRUSTED` | `false` | Deployment-owner acknowledgement for one clean HTTPS external S3 origin; never expose this through connector configuration |
+| `GOVOPLAN_STATE_PROFILE` | `local` | Selects `local`, one-host `host-shared`, or multi-host `shared` state validation |
+| `FILE_UPLOAD_MAX_BYTES` | 50 MiB | Direct-upload and extracted archive-member maximum |
+| `FILE_UPLOAD_ZIP_MAX_BYTES` | 250 MiB | Compressed archive request maximum (legacy name retained for compatibility) |
+| `FILE_ARCHIVE_MAX_ENTRIES` | 10,000 | Maximum declared archive entries |
+| `FILE_ARCHIVE_MAX_EXPANDED_BYTES` | 2 GiB | Maximum expanded archive bytes |
+| `FILE_ARCHIVE_MAX_EXPANSION_RATIO` | 100 | Maximum expanded-to-compressed ratio |
+| `FILE_ARCHIVE_PREVIEW_TTL_SECONDS` | 1,800 | Lifetime of the sealed archive preview token |
 | `MASTER_KEY_B64` | development fallback only | Encrypts database-managed connector secrets |
 
 The local backend is the operational baseline. It resolves every storage key
@@ -407,16 +564,19 @@ under the configured root and rejects escape attempts. Fallback roots support a
 controlled storage-root migration: new writes go to the primary root while
 reads can still find older objects.
 
-The S3 managed-storage adapter currently fails closed before creating a boto3
+The supported installer may provision a deployment-owned Garage service at the
-client because the SDK cannot yet guarantee connection-time DNS/IP pinning and
+exact `http://garage:3900` endpoint and set
-redirect revalidation. Do not select `FILE_STORAGE_BACKEND=s3` for a live Files
+`FILE_STORAGE_S3_DEPLOYMENT_MANAGED=true`. An operator-selected external S3
-deployment until that boundary is implemented and the status above changes.
+backend instead requires a clean HTTPS origin and
+`FILE_STORAGE_S3_ENDPOINT_TRUSTED=true`. Both are deployment authority, not a
+general connector or private-network bypass. Core owns the shared backend
+implementation; Files owns metadata and the Files key namespace.
 
 Multiple API replicas require the same durable blob namespace. Separate local
-container filesystems will produce incomplete reads. Until a pinned shared
+container filesystems will produce incomplete reads. Use `host-shared` with one
-object-storage transport is available, use one durable shared mount or constrain
+durable shared mount only for same-host replicas. Independent hosts require the
-Files traffic to a deployment topology that preserves one consistent local
+`shared` profile with external S3, PostgreSQL, Redis, a stable installation id,
-root.
+and one immutable module composition.
 
 ### Connector egress
 
@@ -441,15 +601,29 @@ The built-in HTTP transport:
 - bounds structured responses to 16 MiB and file transfers to 512 MiB by
   default.
 
+The S3 SDK adapter applies the same socket rule to every botocore pool selected
+for a retry, redirect, discovered endpoint, or virtual-host bucket alias. The
+original authority remains in the request and TLS SNI/certificate check. S3
+connector clients use no outbound proxy and never discover ambient AWS
+credentials: configure both access and secret keys on the governed profile, or
+use an anonymous profile for a public source.
+
+The SMB adapter owns a separate connection cache and replaces smbprotocol's TCP
+factory process-wide with the stricter pinned socket. Initial peers, reconnects,
+server aliases, domain-controller connections, and DFS referral targets therefore
+pass the same policy at connection time. Signing is required by default; enable
+SMB encryption on the profile where the server supports it.
+
 Override the connector response limits with
 `GOVOPLAN_CONNECTOR_MAX_STRUCTURED_RESPONSE_BYTES` and
 `GOVOPLAN_CONNECTOR_MAX_FILE_TRANSFER_BYTES`. The smaller applicable limit wins
 when an import is also subject to `FILE_UPLOAD_MAX_BYTES`.
 
-Never work around a pinning failure by adding a raw IP, disabling TLS, or
+Never work around a connector pinning failure by adding a raw IP, disabling TLS,
-enabling private networks. SMB may redirect through DFS, and S3 SDKs may perform
+or enabling private networks. A failure means the peer policy rejected an actual
-their own redirects or endpoint discovery; both remain disabled even for an IP
+connection destination or the installed SDK no longer exposes the verified
-literal until every connection peer can be governed.
+transport seam. The separately configured platform S3 backend is trusted only by
+the deployment owner and is not selectable by a user or connector profile.
 
 ### Backup and restore
 
@@ -459,13 +633,113 @@ include:
 - Files database rows, including asset/version/blob relationships, shares,
   connector settings, and campaign attachment-use evidence;
 - every object below `FILE_STORAGE_LOCAL_ROOT` and any still-used fallback root;
+  or the complete S3 bucket/prefix and version/lifecycle evidence for an S3
+  backend;
 - the exact `MASTER_KEY_B64` needed to decrypt retained connector credentials;
 - deployment-owned connector profile files, referenced CA bundles, and secret
   environment configuration where those definitions are in use.
 
-There is no Files backup/restore API and no automated blob integrity or orphan
+There is no Files backup/restore API. Use a write quiesce or coordinated
-reconciliation job. Use a write quiesce or coordinated snapshots so database
+snapshots so database references and objects represent the same recovery point.
-references and objects represent the same recovery point.
+The integrity API verifies a restored set, but it does not replace a coordinated
+backup.
+
+Operators normally use **Administration > File integrity**. The equivalent API
+creates a scan with `POST /api/v1/files/integrity/scans`, then calls
+`POST /api/v1/files/integrity/scans/{scan_id}/run` with the scan's current
+`expected_revision` until it reports `completed`. Each call advances at most
+the persisted batch size, so a stopped operator or worker can resume from the
+committed blob/object cursors. Concurrent or stale actions receive `409` before
+the storage backend is invoked; reload the scan and inspect the newer state.
+
+Findings distinguish:
+
+- `missing`: metadata references an absent object;
+- `size_mismatch` or `checksum_mismatch`: bytes do not match immutable blob
+  metadata and the blob is quarantined;
+- `orphan_object`: an object exists in the tenant Files prefix without a
+  corresponding blob row.
+
+Missing or corrupt blobs fail closed for ordinary downloads and Campaign
+attachment materialization. After restoring the expected bytes, use the finding
+`recheck` action with its current `expected_revision`. Orphan cleanup starts
+with a dry-run preview and requires separate destructive confirmation. The
+confirmation reuses the finding revision from that preview, rechecks that no
+database reference exists, remains scoped to the scanned tenant prefix, and is
+idempotent. Both applied and dry-run actions emit audit evidence. A shared
+reference blocks orphan cleanup. Managed-asset purge separately enforces asset
+retention, legal hold, active shares, and retained Campaign/Form evidence before
+releasing a blob for fenced garbage collection.
+
+### Recovery ledger for object effects
+
+On PostgreSQL, every managed blob creation or integrity repair starts a Core
+recovery operation in an independent committed transaction before Files
+protects or writes bytes. The operation records tenant/blob identifiers, an opaque object
+locator or locator digest, semantic SHA-256/size evidence, the recovery mode,
+and a distributed lease fence. It never records file contents, ZIP passwords,
+connector credentials, or a newly uploaded filename. New object keys are opaque;
+legacy filename-bearing keys remain readable but repair operations record only
+their digest and recover through the blob ID.
+
+SQLite is a supported local-development database but permits only one writer.
+Files therefore uses an explicit reduced-durability mode there: recovery intent
+and its lease are written in the caller transaction, while a process-local
+fence prevents competing effects in the same runtime. Commit makes the intent
+durable before independent verification. A handled rollback reconstructs a
+durable recovery operation and verifies compensation or forward completion.
+A hard process loss before commit can leave an object without a surviving
+ledger row, so SQLite is not a production recovery profile; after such a loss,
+run a complete Files integrity scan and reconcile every reported orphan before
+resuming writes. PostgreSQL retains the independent pre-effect durability
+guarantee.
+
+The Files business transaction then creates or updates the blob, version, and
+asset rows. Its actual SQLAlchemy commit or rollback settles every pending
+operation:
+
+- commit reloads the blob through an independent session and streams the object
+  to verify its stored-byte SHA-256 and size before recording success;
+- rollback deletes only a newly reserved object after independently proving
+  that no `FileBlob` references it, then records verified compensation;
+- a repaired existing object is forward-completed only when its identity,
+  envelope, semantic evidence, and stored bytes all match;
+- missing or mismatched bytes quarantine a committed blob and leave the
+  operation `recovery_required`; an unavailable probe remains
+  `outcome_unknown` rather than becoming an ordinary upload failure.
+
+Applied orphan cleanup has its own forward-recovery operation. The database
+reference check and tenant-prefix check happen before deletion; object absence
+and the durable finding state are verified afterward. If the caller transaction
+rolls back after deletion, Files may forward-complete only that existing
+finding after rechecking that the key is still unreferenced.
+
+Hard purge uses an irreversible Core recovery plan with an approval reference
+and a tenant-wide purge lease. The preview hash binds target IDs, deletion
+state, lifecycle revisions, holds, deadlines, blockers, and blob IDs. Asset and
+version removal plus audit evidence commit together. Released objects are not
+deleted by that transaction. Blob GC later takes `files:blob:<tenant>:<blob>`,
+the same distributed resource fence as upload/repair, locks and rechecks the
+blob, and verifies object absence before removing metadata. Provider or database
+ambiguity remains recovery-required or outcome-unknown in Ops.
+
+Explicit S3 connector writes use forward recovery because the provider cannot
+join the database transaction. The durable request contains only tenant,
+profile, opaque target digest, content SHA-256/size, and expected revision. A
+conditional create or overwrite is followed by a metadata probe for the exact
+content and operation markers. Never retry an unresolved target with another
+request key; reconcile the owning Files operation from provider evidence first.
+
+Archive preview and confirmation use bounded process-local temporary staging.
+Staging is not authoritative and is removed on every handled exit; extracted
+members enter the same per-blob recovery boundary as direct uploads. A hard
+process loss may leave a temporary OS file for normal host temporary-file
+cleanup, but cannot make that staging path a managed Files object.
+
+Use the Ops recovery-operation view to inspect `files` operations. Do not retry
+a busy or unresolved blob or connector path blindly: first verify the FileBlob
+row and object hash, or the remote request/content markers and revision, plus
+any Encryption envelope named by a managed blob.
 
 After restore:
 
@@ -493,9 +767,10 @@ Legacy external secret references are detached and identified as non-owned;
 Files never calls a provider delete operation for them.
 
 The retirement executor drops database tables but does not delete corresponding
-objects from the configured blob backend. Operators must include those orphaned
+objects from the configured blob backend. Operators must include those objects
-objects in the approved retention/destruction plan. Validate the installer
+in the approved retention/destruction plan, report them with an integrity scan,
-snapshot and independent blob backup before retirement.
+and explicitly approve cleanup. Validate the installer snapshot and independent
+blob backup before retirement.
 
 ### Operational signals
 
@@ -506,7 +781,7 @@ Use these symptoms as routing hints:
 | `Stored object does not exist` | Database/blob restore mismatch, wrong root, or missing shared storage |
 | `Stored secret cannot be decrypted` | Wrong or rotated `MASTER_KEY_B64` |
 | Private/non-public endpoint blocked | Deployment-wide egress policy is public-only or DNS returned a forbidden answer |
-| SDK cannot pin redirects/referrals | Expected fail-closed S3/SMB boundary, not a transient connector outage |
+| SDK peer-pinning seam is unavailable | Optional S3/SMB SDK is incompatible; keep access fail-closed and validate the supported dependency range before upgrade |
 | Connector response exceeds limit | Remote payload exceeds connector or upload limit |
 | Profile is not visible | Scope, disabled state, campaign access, or policy mismatch |
 | Group removal is vetoed | The group still owns a file/folder/connector space or is a share target |
@@ -522,11 +797,26 @@ public HTTP API. They must not import Files ORM models or storage helpers.
 | --- | --- |
 | `files.access` (`0.1.6`) | Explain resource access provenance for managed files, explicit folders, and virtual folders |
 | `files.campaign_attachments` (`0.1.6`) | Resolve managed attachment matches, prepare frozen campaign snapshots, annotate built messages, share assets with a campaign, and record/mark exact attachment use |
+| `records.source.files` (`1.0.0`) | Recheck current Files access and resolve one exact, integrity-approved managed file version for Records filing |
+| `forms_runtime.evidence.files` (`1.0.0`) | Issue a one-time managed attachment grant and re-verify the exact file/version/checksum at Form submission |
 
 Files requires Core principal resolution and permission evaluation. Campaign is
 an optional dependency; when installed, Files consumes the optional
 `campaigns.access` interface to verify campaign existence and access. Missing
 optional Campaign support fails explicitly rather than bypassing the check.
+Records is also optional. When enabled, the source capability returns the
+requested `FileVersion` identity, path snapshot, content metadata, SHA-256,
+integrity/protection state, and launch link. It rejects mutable aliases,
+cross-tenant requests, missing access, and quarantined or failed blobs. Records
+stores the filing decision; Files continues to own the version and bytes.
+Forms Runtime is optional as well. Its public or authenticated participant never
+receives general Files access through this integration. Forms Runtime asks for a
+purpose-bound grant, Files verifies an active same-tenant user custodian, stores
+only the token digest, accepts one bounded upload, and returns an immutable
+`EvidenceReference`. Draft save and final submit independently recheck the exact
+Form instance/revision, grant, asset, version, checksum, deletion state, and
+integrity state. An idempotent replay returns the existing grant without
+reproducing its bearer secret.
 
 ### API families
 
@@ -535,17 +825,32 @@ All routes below are under `/api/v1/files`.
 | Area | Routes |
 | --- | --- |
 | Spaces and content | `GET /spaces`, `GET /`, `GET /folders`, `GET /delta` |
-| Upload and folders | `POST /upload`, `POST /upload-zip`, `POST /folders`, `POST /folders/delete` |
+| Upload and folders | `POST /upload`, `POST /upload-zip` (compatibility), `POST /archive-preview`, `POST /archive-confirm`, `POST /folders`, `POST /folders/delete` |
 | File access | `GET /{file_id}`, `GET /{file_id}/download`, `DELETE /{file_id}`, `POST /bulk-delete` |
 | Organization | `POST /bulk-rename`, `POST /transfer`, `POST /archive.zip`, `POST /resolve-patterns` |
 | Sharing | `POST /{file_id}/shares`, `POST /bulk-shares` |
-| Connector spaces | `GET/POST /connector-spaces`, `PATCH/DELETE /connector-spaces/{space_id}` |
+| Connector spaces | `GET/POST /connector-spaces`, `PATCH/DELETE /connector-spaces/{space_id}`, `POST /connector-spaces/{space_id}/restore`, `POST /connector-spaces/{space_id}/sync`, `POST /connector-spaces/{space_id}/write-back` |
 | Connector catalog/discovery | `GET /connectors/providers`, `POST /connectors/discover` |
 | Connector profiles | `GET/POST /connectors/profiles`, `GET/PATCH/DELETE /connectors/profiles/{profile_id}` |
 | Browse/import/sync | `GET /connectors/profiles/{profile_id}/browse`, `POST /connectors/profiles/{profile_id}/import`, `POST /connectors/profiles/{profile_id}/sync` |
+| Restore and lifecycle | `POST /assets/{file_id}/restore`, `POST /folders/restore`, `PATCH /{file_id}/lifecycle` |
+| Governed erasure | `POST /purge/preview`, `POST /purge/execute`, `POST /purge/blobs` |
 | Credentials | `GET/POST /connectors/credentials`, `GET/PATCH/DELETE /connectors/credentials/{credential_id}` |
 | Policy | `GET/PUT /connectors/policies/{scope_type}`, `POST /connector-policy/evaluate` |
 | Incremental connector settings | `GET /connectors/settings/delta` |
+| Form evidence | `POST /form-evidence/upload` with a short-lived `X-Form-Evidence-Token` issued by Forms Runtime |
+
+The Files workspace exposes **Remove space** only for connector spaces and only
+to actors with file-organization authority over the owning user
+or group space. Confirmation explains the exact boundary: removal soft-deletes
+the local connector-space definition and makes that virtual view disappear. It
+does not mutate or delete remote provider content, previously imported managed
+files, their metadata or shares, the connector profile, credentials, or remote
+object references. Those retained objects therefore do not block removal and
+the connector location can be linked again later. User and group managed spaces
+are intrinsic ownership scopes rather than removable records, so they never
+offer this action. A missing, already removed, cross-tenant, or inaccessible
+connector space fails through the backend lookup and owner-access checks.
 
 Consumers should use cursor/watermark contracts instead of assuming an
 unbounded complete list. The default full-list page size is 500 and public page
@@ -580,21 +885,32 @@ Files baseline indiscriminately.
 - Owner and share checks protect individual resources after operation-scope
   checks.
 - Logical paths reject traversal and storage keys cannot escape the local root.
-- Upload, ZIP extraction, connector response, and S3 stream code use bounded
+- Upload, archive extraction, connector response, and S3 stream code use bounded
-  reads; encrypted ZIP archives are rejected.
+  reads. Archive previews are sealed and short-lived; ZIP passwords are
+  request-only.
+- Public Form evidence uploads require a custom-header bearer grant that is
+  stored only as SHA-256, expires after at most 15 minutes, is bound to one
+  exact submission and user custodian, and can create only one managed file.
 - Connector HTTP sockets use connection-time DNS/IP validation and pinning,
   redirects are refused, and unsafe SDK transports fail before client creation.
 - Database-managed connector passwords/tokens are encrypted; responses redact
   secrets and deployment references.
 - API metadata is recursively checked for secret-like values.
 - Downloads use sanitized attachment filenames.
-- SHA-256 and byte size are recorded for every blob/version.
+- Plaintext semantic and stored-byte SHA-256/size evidence are recorded for
+  every protected blob; they are identical for unprotected blobs.
+- Upload and archive-confirm APIs can select an Encryption vault. Protected
+  writes and reads fail closed if the optional Encryption capability is absent.
+- Managed-object writes and applied orphan cleanup start lease-fenced Core
+  recovery operations before their physical effects; terminal success and
+  compensation require independent database and object checks.
 
 The module does **not** currently provide malware scanning, content disarm and
-reconstruction, a file-type allowlist, per-user quota, at-rest encryption for
+reconstruction, a file-type allowlist, per-user quota, automatic encryption
-local blob bytes, or a dedicated preview sandbox. Deployments that require
+policy assignment, client E2EE, or a dedicated preview sandbox. The optional
-these controls must supply them outside Files until explicit module contracts
+server-envelope profile protects selected managed blob bytes at rest but remains
-exist.
+server-decryptable. Deployments that require the other controls must supply
+them outside Files until explicit module contracts exist.
 
 ### Provenance
 
@@ -616,7 +932,7 @@ events. External metadata is provider/user input and is not a digital signature.
 Files records canonical audit events for:
 
 - connector discovery attempts, before the attempted external I/O;
-- connector imports and manual syncs;
+- connector imports, selected-file syncs, and bounded folder sync summaries;
 - download/archive access to connector-originated managed files;
 - immediate connector profile and credential deletion/scrubbing;
 - credential scrubbing during destructive module retirement.
@@ -624,6 +940,10 @@ Files records canonical audit events for:
 Connector audit details include the managed asset/version/blob, checksum, size,
 operation, source revision, and provenance where applicable. Deletion audit
 details name secret/reference kinds but never the secret values.
+`files.connector.folder_synced` additionally records the connector profile,
+provider, library/root, managed target, traversal mode, truncation state, typed
+counts, and bounded per-item source/target/action references. It does not embed
+downloaded bytes, credentials, or transport content.
 
 Assets, folders, shares, profiles, credentials, policies, and connector spaces
 also feed Core's incremental change sequence for UI synchronization. A change
@@ -647,6 +967,9 @@ The word "delete" has different meanings by object type:
 | Connector profile | Immediately disables the tombstone and clears credential links/material/references and private metadata |
 | Legacy `secret_ref` | Detached and audited as an unowned external reference; no provider deletion is attempted or claimed |
 | Module retirement | Scrubs/audits credential material, then drops Files database tables; blob-backend cleanup is an operator responsibility |
+| Governed hard purge | Removes only preview-matched, soft-deleted asset/version rows without active retention, legal hold, active shares, Campaign evidence, or Form evidence |
+| Blob garbage collection | Deletes an exact managed object only after a fresh zero-reference check under the shared blob fence; metadata follows verified object absence |
+| Connector write-back | Explicit conditional S3 create/overwrite only; automatic provider delete, rename, move, and ACL propagation remain disabled |
 
 Credential/profile scrubbing and its audit event use the same database
 transaction. If audit creation fails, the deletion rolls back. Repeating a
@@ -654,15 +977,19 @@ delete against an already scrubbed tombstone does not recreate secret evidence.
 
 ### Retention boundary
 
-File versions and blobs are effectively retained indefinitely today. Although a
+Each asset has an enforceable retained-until value, legal-hold flag, reason, and
-blob has `ref_count` and `retained_until` fields, no complete retention-policy,
+optimistic lifecycle revision. These controls block hard purge; they do not
-legal-hold, hard-purge, or garbage-collection service enforces them. There is
+automatically schedule it. Restore, purge preview/execute, and blob GC are
-also no supported user restore endpoint for soft-deleted assets/folders and no
+explicit authorized operations. Blob `retained_until` remains an additional
-automated orphan-object reconciliation.
+storage-level safeguard and must also have expired before automated collection
+is introduced. Campaign and Form evidence are hard blockers rather than
+silently cascaded references.
 
-Do not promise erasure, timed retention, legal hold, or self-service recovery
+Do not equate soft deletion with erasure. Erasure is complete only after the
-from the current soft-delete behavior. Those require an explicit, auditable
+approved purge removes asset/version metadata, bounded GC verifies that no
-retention/purge design that preserves campaign and other evidence references.
+version references the blob and removes its bytes/metadata, and any owning
+Encryption retention/key-custody consequence has been handled under that
+module's policy.
 
 ## Acceptance scenarios
 
@@ -675,12 +1002,14 @@ her personal or Finance space, then the asset has the selected owner and tenant.
 Given Bob is outside Finance and has no share or Files admin permission, the
 same asset ID must not make the asset readable to Bob.
 
-### Safe ZIP import
+### Safe archive import
 
-Given a ZIP member contains `../../secret.txt`, is encrypted, exceeds 50 MiB,
+Given an archive member contains `../../secret.txt`, is a link or special
-or pushes actual extracted bytes above 250 MiB, unpacking must fail without a
+filesystem object, exceeds 50 MiB, pushes actual expanded bytes above 2 GiB, or
-committed managed asset. A valid archive must preserve normalized relative paths
+exceeds the 100:1 expansion ratio, preview or confirmation must fail without a
-below the chosen logical folder.
+committed managed asset. A password-protected ZIP requires the correct
+request-only password. A valid confirmation must match its unexpired preview
+token and preserve normalized relative paths below the chosen logical folder.
 
 ### Explicit conflicts
 
@@ -699,9 +1028,11 @@ bypass the same rule.
 ### Pinned connector transport
 
 Given public-only mode and DNS returns any private address, the connection must
-be rejected before a socket opens. Given private mode, the HTTP connection must
+be rejected before a socket opens. Given private mode, every HTTP, S3, and SMB
-still use the validated address and reject redirects. SMB and S3 must fail
+connection must still use an address from the answer validated for that exact
-before their SDK clients connect until all SDK-managed peers can be pinned.
+attempt. Botocore retries, redirects, endpoint discovery, and aliases, plus SMB
+reconnects and DFS referrals, must pass through the pinned factories. A changed
+or unsupported SDK seam must fail before a usable client/session is returned.
 
 ### Imported evidence and sync
 
@@ -710,6 +1041,13 @@ must carry source identity, revision when available, current version, checksum,
 and size. Re-syncing identical bytes must return `unchanged`; changed bytes must
 create a higher version while preserving the previous version.
 
+Given a permitted linked connector folder contains changed, unchanged,
+colliding, and policy-denied files, a manual folder sync must preserve remote
+relative paths, commit the allowed siblings, return one typed outcome for every
+discovered/skipped item, and emit a matching summary audit event. Reaching the
+configured file bound must mark the run truncated rather than imply complete
+coverage.
+
 ### Campaign freeze
 
 Given a campaign snapshot selected version V1, when the managed asset later
@@ -735,20 +1073,22 @@ returning different content or credentials.
 
 | Area | Implemented now | Planned or explicitly outside the current boundary |
 | --- | --- | --- |
-| Managed storage | Local durable-root backend, fallback read roots, tenant blob deduplication, checksums | Operational S3 after pinned SDK transport ([#34](https://git.add-ideas.de/add-ideas/govoplan-files/issues/34)); automated integrity/orphan reconciliation ([#36](https://git.add-ideas.de/add-ideas/govoplan-files/issues/36)) |
+| Managed storage | Core local/S3 backend, exact managed-Garage or explicitly trusted HTTPS external S3, state-profile validation, fallback local read roots, tenant blob deduplication, checksums, bounded resumable integrity scans, quarantine, dry-run-first orphan cleanup, and Core-ledger verification/forward recovery | Scheduled scan execution and deployment-specific S3 HA/backup automation |
-| Upload | Bounded direct upload, drag-and-drop UI, ZIP spool/extract limits, explicit conflicts | Malware scanning, quotas, type policy, resumable/chunked upload |
+| Upload | Bounded direct upload, drag-and-drop UI, archive preview/selective extraction, password-protected ZIP support, explicit conflicts, opaque new object keys, and rollback compensation | Malware scanning, quotas, type policy, resumable/chunked upload |
-| Organization | Folders, bulk rename preview/apply, move/copy, drag-and-drop, ZIP download, pattern resolution | General file-history UI and user-driven append-version/restore |
+| Organization | Folders, bulk rename preview/apply, move/copy, drag-and-drop, ZIP download, pattern resolution, and API restoration preserving versions/provenance | General file-history UI and user-driven append-version UI |
-| Sharing | User/group/tenant/campaign grant or permission update through API; campaign linkage display | Share revocation/expiry and complete general share-management UI ([#37](https://git.add-ideas.de/add-ideas/govoplan-files/issues/37)) |
+| Sharing | User/group/tenant/campaign grants, expiry, idempotent revocation, searchable share-management UI, and campaign linkage display | Richer policy-driven share lifecycles |
-| Deletion/retention | Soft-delete assets/folders/spaces; immediate audited connector-secret scrubbing | File restore API, hard purge, retention policy, legal hold, and blob GC ([#38](https://git.add-ideas.de/add-ideas/govoplan-files/issues/38)) |
+| Deletion/retention | Soft-delete and restore assets/folders/spaces; optimistic retention and legal-hold controls; preview-bound, approval-referenced hard purge; reference-checked blob GC; immediate audited connector-secret scrubbing | Automatic time-based purge scheduling and richer lifecycle administration UI |
+| Privacy requests | Tenant-scoped bounded DSAR metadata search; retained/manual/revoke/detach planning; idempotent share revocation and mutable actor-reference detachment; explicit separation from byte purge | Content-specific automated redaction and policy-specific approval remain manual or belong to the owning process |
 | Connector governance | Scoped profiles/credentials/policies, effective source explanation, separate credentials, linked user/group spaces | Provider-owned external secret lifecycle; API `secret_ref` remains rejected |
-| HTTP connectors | Pinned, bounded, no-redirect Seafile and WebDAV/Nextcloud browse/import/manual sync | Background/folder sync, remote mutation, long-running transfer workers |
+| HTTP connectors | Pinned, bounded, no-redirect Seafile and WebDAV/Nextcloud browse/import, selected-file sync, and folder sync | Background sync, remote mutation, long-running transfer workers |
-| SMB and S3 connectors | Provider descriptors and browse/import logic | Live access only after SDK connections plus DFS referrals/redirects are validated and pinned ([SMB #35](https://git.add-ideas.de/add-ideas/govoplan-files/issues/35), [S3 #34](https://git.add-ideas.de/add-ideas/govoplan-files/issues/34)) |
+| SMB and S3 connectors | Provider descriptors, browse/import/manual sync, pinned SDK transports, redirect/retry/referral transport-contract tests, and explicit conditional S3 write-back with Core-ledger recovery | Live topology smoke evidence, provider-specific OAuth, additional provider writes, and background indexing remain separate deployment or connector-module concerns |
 | Other providers | Reserved SharePoint/OneDrive keys and NFS/local descriptors | Graph/OAuth/provider paging, NFS deployment integration, DMS connectors |
-| Connector spaces | User/group link, browse, manual selected-file sync, edit/disable/delete | Background sync, remote writes/deletes, full conflict-reporting jobs |
+| Connector spaces | User/group link, browse, bounded manual file/folder sync, edit/disable/delete/restore, read-only default, and opt-in S3 two-way mode | Background sync and automatic remote delete/rename/move/ACL propagation |
-| Profile capabilities | Stored and displayed | Enforce capability flags as an independent operation gate |
+| Profile capabilities | Stored, displayed, and enforced for explicit connector write-back | Broader provider-specific capability negotiation |
 | Audit | Connector discovery/import/sync/access and connector deletion; campaign exact-use evidence | Dedicated canonical audit events for every ordinary Files mutation |
 | Preview | File metadata and attachment download | Dedicated safe content-preview service |
 | Campaign | Stable capability-based frozen attachments and sent-use evidence | Campaign-specific process state remains in Campaign |
+| Forms Runtime | One-time managed attachment grants plus exact-version, checksum, deletion, tenant, submission, and integrity verification | Malware scanning and advanced/qualified signature providers remain separate assurance depth |
 | Collaboration | Governed input/output snapshots | Co-editing, comments, review, presence, locks, and semantic document versions belong to Documents/workflow/provider modules |
 
 ## Release and change checklist
@@ -764,11 +1104,20 @@ Before releasing Files:
 5. Exercise an allowed and denied owner/share path.
 6. Exercise upload, ZIP bounds, conflict handling, download, and soft deletion.
 7. Exercise connector policy explanation and one pinned HTTP provider where
-   configured; verify S3/SMB still fail closed.
+   configured; verify the deployment-managed or trusted external S3 backend if
+   selected, and verify one configured S3 and SMB connector while recording the
+   actual target topology and private-network policy.
 8. Verify credential deletion scrubs dependents and produces audit evidence.
 9. Verify a campaign attachment snapshot still identifies its exact version and
    checksum after the current file changes.
-10. Update the implemented/planned table whenever a boundary changes.
+10. Exercise a committed upload, a rolled-back upload, object tamper detection,
+    and applied orphan cleanup; inspect their `files` operations in Ops.
+11. Exercise a Form evidence grant, token replay, wrong-submission reference,
+    expired token, unsupported media type, and quarantined-file rejection.
+12. Exercise a Files DSAR search/plan, repeat an approved reversible action, and
+    confirm retained file content can only be erased through the separate purge
+    authority and recovery path.
+13. Update the implemented/planned table whenever a boundary changes.
 
 ## Related documents
 

docs/GENERATED_ARTIFACT_STORE.md

@@ -0,0 +1,23 @@
+# Generated Artifact Store
+
+Files implements Core's optional `files.artifact_store` capability for modules
+that generate deterministic output without owning file storage.
+
+The producer supplies bytes, filename, content type, destination folder,
+optional idempotency key, and bounded non-secret provenance. Files applies the
+actor's `files:file:upload` permission, tenant/user ownership, path rules,
+versioning, configured blob backend, and conflict behavior. An idempotency key
+is represented as source provenance so an unchanged retry does not create an
+unrelated file version.
+
+The shared Files session owns finalization. Before a new managed object is
+written, Files commits a lease-fenced Core recovery operation containing only
+identifiers and digests. The caller's eventual session commit independently
+verifies both `FileBlob` metadata and stored bytes; rollback compensates only an
+unreferenced key. Producers must therefore complete the supplied transaction
+normally and must not bypass or replace Files session lifecycle handling.
+
+The response contains only file/version identifiers, display path, media type,
+size, digest, and storage provenance. Producers must not put credentials,
+tokens, or rendered plaintext into metadata. Storing an artifact proves Files
+accepted it; it does not prove printing, mailing, or any other external effect.

docs/INTERFACE_PATTERN_MIGRATION.md

@@ -0,0 +1,55 @@
+# Files Interface Pattern Migration
+
+This inventory records the Files-owned part of the GovOPlaN interface pattern
+language. Core owns the shell and shared components; Files owns the composition
+and consequences described here.
+
+## Surface inventory
+
+| Surface | Primary task | Archetype | Consequence | Pattern evidence |
+| --- | --- | --- | --- | --- |
+| `/files` space and folder panes | Browse managed and connected content without losing location | Directory/explorer | Low for navigation; medium for exposing filenames and provenance | Full-height two-pane workspace, bounded panes, stable selection and contextual Help Center link |
+| `/files` toolbar and property filters | Find and act on the current selection | Explorer actions and local filtering | Medium for upload, move, copy, share and synchronization; high for delete | Actions remain beside the affected list, disabled controls explain permission/state/selection blockers, destructive work uses `ConfirmDialog` |
+| Upload/archive, transfer, rename and connector-import dialogs | Supply and review one bounded change | Adaptive create/edit or guided import | Medium to high because files, paths and external bytes change | Shared `Dialog`, `FileDropZone`, validation, conflict review, unsaved inputs and explicit confirmation |
+| File share dialog | Inspect and change access | Review/decision | High because another actor gains access | Shared dialog, access explanation, stable row actions and destructive confirmation |
+| System/tenant/group/user connector surfaces | Compare connections, credentials and effective policy | Administration/configuration | High because endpoints, secrets and inherited policy control external access | Shared `ConnectionTree`, adaptive forms, `ActionBlockerHint`, policy provenance and contextual admin help |
+| Connection and credential editors | Create or edit one governed endpoint or secret | Adaptive create/edit | High because a saved change may enable remote access | Relevant fields only, typed credential controls, discovery/test, advanced compatibility section, unsaved-change guard and disabled-save reasons |
+| Connector policy card | Narrow inherited connector access | Effective-policy editor | High because deny/allow changes affect lower scopes | Typed reference selectors, deny precedence warning, effective source evidence and permission blocker |
+| `files.widget.spaces` | See available managed/connected spaces and open Files | Dashboard widget | Low; names and provider state may still be sensitive | Shared loading/alert/status components, bounded item count, permission-filtered contribution |
+| Files chooser capability used by another module | Select a managed snapshot without importing Files internals | Directory chooser | Medium because the exact selected version becomes another module's input | Shared dialog/confirmation, capability boundary and exact file/version evidence |
+
+## State and consequence contract
+
+- Loading, errors, success, empty results, access explanations and confirmation
+  use Core components. Files does not reproduce the application shell.
+- A connector that comes from deployment settings remains visible but read-only;
+  its action explains that bootstrap configuration and a restart are required.
+- Missing permission, target, selection, endpoint, or compatible provider is an
+  explained disabled state. It is not represented only by color or absence.
+- Provider metadata JSON is an expert compatibility escape hatch inside the
+  collapsed shared advanced-options component. Ordinary connector setup uses
+  typed provider, endpoint, credential, capability and policy controls.
+- Endpoint discovery and credential tests are explicit and report their result;
+  they do not save the draft. Save remains the only committing action.
+- Connector/profile disable and managed-file delete remain confirmed actions and
+  state their immediate effect. Soft deletion must not be described as purge.
+- External connector data, paths and credential references are rendered only in
+  already-authorized administration or explorer contexts. Secret values are
+  never returned for rendering.
+
+## Accessibility and responsive evidence
+
+Shared `Dialog` owns focus entry, Escape handling and focus return. Form and
+toolbar DOM order is the keyboard order; disabled-action tooltips are themselves
+focusable and expose the reason. The connector form uses semantic sections and
+labels, status is textual as well as colored, and result alerts are announced by
+the shared alert component. The explorer collapses to one column below 1050 px;
+connector forms and action rows collapse below 760 px while preserving source
+order. Long provider choices scroll inside the segmented control rather than
+expanding the page.
+
+The focused structural test guards these contracts, optional-module boundaries,
+confirmation, contextual help, advanced-only JSON and responsive rules. Core's
+TypeScript build, structural localization audit, module-permutation suite and
+full-product bundle check provide the integration gates.
+

package.json

@@ -1,6 +1,6 @@
 {
   "name": "@govoplan/files-webui",
-  "version": "0.1.9",
+  "version": "0.1.20",
   "private": true,
   "type": "module",
   "main": "webui/src/index.ts",
@@ -19,14 +19,14 @@
     "LICENSE"
   ],
   "peerDependencies": {
-    "@govoplan/core-webui": "^0.1.9",
+    "@vitejs/plugin-react": "^5.2.0",
-    "lucide-react": "^1.23.0",
+    "vite": "^7.3.6",
-    "react": "^19.0.0",
-    "react-dom": "^19.0.0",
-    "react-router-dom": "^7.1.1",
-    "@vitejs/plugin-react": "^4.3.4",
     "typescript": "^5.7.2",
-    "vite": "^6.0.6"
+    "react": ">=19.2.7 <20",
+    "react-dom": ">=19.2.7 <20",
+    "react-router": ">=8.3.0 <9",
+    "lucide-react": "^1.23.0",
+    "@govoplan/core-webui": "^0.1.18"
   },
   "peerDependenciesMeta": {
     "@govoplan/core-webui": {

pyproject.toml

@@ -4,15 +4,16 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "govoplan-files"
-version = "0.1.9"
+version = "0.1.20"
 description = "GovOPlaN files module with backend and WebUI integration."
 readme = "README.md"
 requires-python = ">=3.12"
 license = { file = "LICENSE" }
 authors = [{ name = "GovOPlaN" }]
 dependencies = [
-  "govoplan-core>=0.1.9",
+  "govoplan-core>=0.1.20",
   "defusedxml>=0.7,<1",
+  "pyzipper>=0.3.6,<1",
   "python-multipart>=0.0.31,<1",
 ]
 

src/govoplan_files/backend/capabilities.py

@@ -5,13 +5,32 @@ import binascii
 from typing import Any
 
 from sqlalchemy import or_
+from sqlalchemy.orm import Session
 
+from govoplan_core.audit.logging import audit_from_principal
+from govoplan_core.auth import ApiPrincipal, has_scope
 from govoplan_core.core.access import AccessDecisionProvenance, PrincipalRef
 from govoplan_core.core.files import FileAccessProvider
+from govoplan_core.core.files import (
+    ManagedTabularFile,
+    ManagedTabularFileAccessError,
+    ManagedTabularFileContent,
+    ManagedTabularFileNotFoundError,
+    ManagedTabularFileProvider,
+    ManagedTabularFileUnavailableError,
+    ManagedTabularFileValidationError,
+    PostboxFileReferenceRef,
+    PostboxFileReferenceRequest,
+    PostboxFileReferenceProvider,
+    ManagedArtifactRef,
+    ManagedArtifactStore,
+    ManagedArtifactWriteRequest,
+)
 from govoplan_core.core.modules import ModuleContext
 from govoplan_core.security.module_permissions import scopes_grant_compatible
-from govoplan_files.backend.db.models import FileAsset, FileFolder, FileShare
+from govoplan_files.backend.db.models import FileAsset, FileBlob, FileFolder, FileShare, FileVersion
 from govoplan_files.backend.runtime import configure_runtime
+from govoplan_files.backend.storage.common import FileStorageError
 from govoplan_files.backend.storage.campaign_attachments import (
     annotate_built_messages_with_managed_files,
     managed_match_payloads,
@@ -20,8 +39,16 @@ from govoplan_files.backend.storage.campaign_attachments import (
     share_assets_with_campaign,
 )
 from govoplan_files.backend.storage.campaign_usage import record_campaign_attachment_uses_for_jobs
-from govoplan_files.backend.storage.files import current_version_and_blob
+from govoplan_files.backend.storage.files import (
+    create_file_asset,
+    current_version_and_blob,
+    get_asset_for_user,
+    list_recent_assets_for_user,
+    read_asset_version_bytes,
+    sync_file_asset_from_source,
+)
 from govoplan_files.backend.storage.paths import normalize_folder
+from govoplan_files.backend.storage.share_state import effective_file_share_clause
 
 
 VIRTUAL_FOLDER_RESOURCE_PREFIX = "virtual-folder:v1"
@@ -59,6 +86,214 @@ def campaign_capability(context: ModuleContext) -> FilesCampaignCapability:
     return FilesCampaignCapability()
 
 
+class FilesArtifactStore(ManagedArtifactStore):
+    def store_artifact(
+        self,
+        session: object,
+        principal: object,
+        *,
+        request: ManagedArtifactWriteRequest,
+    ) -> ManagedArtifactRef:
+        if not hasattr(session, "query") or not hasattr(session, "flush"):
+            raise TypeError("Files artifact storage requires a SQLAlchemy session.")
+        if not hasattr(principal, "has") or not principal.has("files:file:upload"):
+            raise PermissionError("Managed artifact storage requires files:file:upload.")
+        user = getattr(principal, "user", None)
+        user_id = str(getattr(user, "id", "") or "")
+        tenant_id = str(getattr(principal, "tenant_id", "") or "")
+        if not user_id or not tenant_id:
+            raise PermissionError("Managed artifact storage requires a tenant user principal.")
+        metadata = dict(request.metadata)
+        if request.idempotency_key:
+            metadata["source_provenance"] = {
+                "source_type": "generated_artifact",
+                "connector_id": "files.artifact_store",
+                "provider": str(metadata.get("producer_module") or "platform"),
+                "external_id": request.idempotency_key,
+                "revision": str(metadata.get("output_sha256") or "") or None,
+            }
+            stored, _action, _previous_version_id = sync_file_asset_from_source(
+                session,
+                tenant_id=tenant_id,
+                owner_type="user",
+                owner_id=user_id,
+                user_id=user_id,
+                filename=request.filename,
+                data=request.payload,
+                metadata=metadata,
+                folder=request.folder,
+                content_type=request.content_type,
+                conflict_strategy="rename",
+                is_admin=principal.has("files:file:admin"),
+            )
+        else:
+            stored = create_file_asset(
+                session,
+                tenant_id=tenant_id,
+                owner_type="user",
+                owner_id=user_id,
+                user_id=user_id,
+                filename=request.filename,
+                data=request.payload,
+                folder=request.folder,
+                content_type=request.content_type,
+                description=request.description,
+                metadata=metadata,
+                conflict_strategy="rename",
+                is_admin=principal.has("files:file:admin"),
+            )
+        return ManagedArtifactRef(
+            file_asset_id=stored.asset.id,
+            file_version_id=stored.version.id,
+            filename=stored.version.filename_at_upload,
+            display_path=stored.asset.display_path,
+            content_type=stored.version.content_type or request.content_type,
+            size_bytes=stored.version.size_bytes,
+            sha256=stored.version.checksum_sha256,
+            provenance={
+                "module": "files",
+                "owner_type": stored.asset.owner_type,
+                "managed": True,
+            },
+        )
+
+
+def artifact_store_capability(context: ModuleContext) -> FilesArtifactStore:
+    configure_runtime(registry=context.registry, settings=context.settings)
+    return FilesArtifactStore()
+
+
+class FilesPostboxReferenceService(PostboxFileReferenceProvider):
+    """Resolve Postbox evidence without turning message access into file access."""
+
+    def resolve_postbox_references(
+        self,
+        session: object,
+        principal: object,
+        *,
+        tenant_id: str,
+        requests: tuple[PostboxFileReferenceRequest, ...],
+    ) -> tuple[PostboxFileReferenceRef, ...]:
+        if not hasattr(session, "get"):
+            raise TypeError("Postbox file resolution requires a SQLAlchemy session.")
+        principal_tenant = str(getattr(principal, "tenant_id", "") or "")
+        can_download = bool(
+            hasattr(principal, "has") and principal.has("files:file:download")
+        )
+        user = getattr(principal, "user", None)
+        user_id = str(getattr(user, "id", "") or "")
+        is_admin = bool(
+            hasattr(principal, "has") and principal.has("files:file:admin")
+        )
+        results: list[PostboxFileReferenceRef] = []
+        for request in requests:
+            if principal_tenant != tenant_id:
+                results.append(_unavailable_postbox_reference(request, "tenant_mismatch"))
+                continue
+            if not can_download or not user_id:
+                results.append(
+                    _unavailable_postbox_reference(
+                        request,
+                        "download_permission_missing",
+                    )
+                )
+                continue
+            reference_type = request.reference_type.strip().casefold()
+            version: FileVersion | None = None
+            if reference_type in {"file", "file_asset", "files:file"}:
+                results.append(
+                    _unavailable_postbox_reference(request, "exact_version_required")
+                )
+                continue
+            elif reference_type in {"file_version", "files:file_version"}:
+                version = session.get(FileVersion, request.reference_id)  # type: ignore[attr-defined]
+                asset = (
+                    session.get(FileAsset, version.file_asset_id)  # type: ignore[attr-defined]
+                    if version is not None
+                    else None
+                )
+            else:
+                results.append(
+                    _unavailable_postbox_reference(
+                        request,
+                        "unsupported_reference_type",
+                    )
+                )
+                continue
+            if (
+                asset is None
+                or version is None
+                or asset.tenant_id != tenant_id
+                or version.tenant_id != tenant_id
+                or asset.deleted_at is not None
+            ):
+                results.append(_unavailable_postbox_reference(request, "file_not_found"))
+                continue
+            try:
+                get_asset_for_user(
+                    session,  # type: ignore[arg-type]
+                    tenant_id=tenant_id,
+                    user_id=user_id,
+                    asset_id=asset.id,
+                    is_admin=is_admin,
+                )
+            except FileStorageError:
+                results.append(
+                    _unavailable_postbox_reference(request, "file_access_denied")
+                )
+                continue
+            blob = session.get(FileBlob, version.blob_id)  # type: ignore[attr-defined]
+            if blob is None or blob.tenant_id != tenant_id:
+                results.append(
+                    _unavailable_postbox_reference(request, "file_payload_missing")
+                )
+                continue
+            results.append(
+                PostboxFileReferenceRef(
+                    reference_type=request.reference_type,
+                    reference_id=request.reference_id,
+                    available=True,
+                    reason_code="available",
+                    file_asset_id=asset.id,
+                    file_version_id=version.id,
+                    filename=version.filename_at_upload or asset.filename,
+                    content_type=version.content_type or blob.content_type,
+                    size_bytes=version.size_bytes,
+                    sha256=version.checksum_sha256,
+                    download_path=(
+                        f"/api/v1/files/{asset.id}/versions/{version.id}/download"
+                    ),
+                    provenance={
+                        "module": "files",
+                        "postbox_id": request.postbox_id,
+                        "message_id": request.message_id,
+                        "display_path": asset.display_path,
+                        "exact_version": True,
+                    },
+                )
+            )
+        return tuple(results)
+
+
+def _unavailable_postbox_reference(
+    request: PostboxFileReferenceRequest,
+    reason_code: str,
+) -> PostboxFileReferenceRef:
+    return PostboxFileReferenceRef(
+        reference_type=request.reference_type,
+        reference_id=request.reference_id,
+        available=False,
+        reason_code=reason_code,
+    )
+
+
+def postbox_reference_capability(
+    context: ModuleContext,
+) -> FilesPostboxReferenceService:
+    configure_runtime(registry=context.registry, settings=context.settings)
+    return FilesPostboxReferenceService()
+
+
 class FilesAccessService(FileAccessProvider):
     def explain_resource_provenance(
         self,
@@ -96,7 +331,7 @@ class FilesAccessService(FileAccessProvider):
             .filter(
                 FileShare.tenant_id == asset.tenant_id,
                 FileShare.file_asset_id == asset.id,
-                FileShare.revoked_at.is_(None),
+                effective_file_share_clause(),
                 FileShare.permission.in_(sorted(permission_values)),
                 or_(
                     (FileShare.target_type == "user") & (FileShare.target_id == principal.membership_id),
@@ -197,6 +432,228 @@ def access_capability(context: ModuleContext) -> FilesAccessService:
     return FilesAccessService()
 
 
+class FilesManagedTabularFileService(ManagedTabularFileProvider):
+    """Expose authorized immutable CSV/XLSX versions to optional consumers."""
+
+    def list_tabular_files(
+        self,
+        session: object,
+        principal: object,
+        *,
+        query: str = "",
+        limit: int = 100,
+    ) -> tuple[ManagedTabularFile, ...]:
+        db, api_principal, user_id = _tabular_context(
+            session,
+            principal,
+            required_scope="files:file:read",
+        )
+        normalized_query = str(query or "").strip().casefold()
+        requested_limit = max(1, min(int(limit), 100))
+        assets = list_recent_assets_for_user(
+            db,
+            tenant_id=api_principal.tenant_id,
+            user_id=user_id,
+            limit=min(500, requested_limit * 5),
+            is_admin=has_scope(api_principal, "files:file:admin"),
+        )
+        results: list[ManagedTabularFile] = []
+        for asset in assets:
+            if normalized_query and normalized_query not in (
+                f"{asset.filename} {asset.display_path} {asset.description or ''}"
+            ).casefold():
+                continue
+            try:
+                version, blob = current_version_and_blob(db, asset)
+            except FileStorageError:
+                continue
+            if not _is_tabular_file(version.filename_at_upload, version.content_type):
+                continue
+            results.append(_managed_tabular_file(asset, version, blob))
+            if len(results) >= requested_limit:
+                break
+        return tuple(results)
+
+    def get_tabular_file(
+        self,
+        session: object,
+        principal: object,
+        *,
+        file_asset_id: str,
+        file_version_id: str | None = None,
+    ) -> ManagedTabularFile | None:
+        db, api_principal, user_id = _tabular_context(
+            session,
+            principal,
+            required_scope="files:file:read",
+        )
+        try:
+            asset = get_asset_for_user(
+                db,
+                tenant_id=api_principal.tenant_id,
+                user_id=user_id,
+                asset_id=file_asset_id,
+                is_admin=has_scope(api_principal, "files:file:admin"),
+            )
+        except FileStorageError:
+            return None
+        if file_version_id:
+            version = db.get(FileVersion, file_version_id)
+            if (
+                version is None
+                or version.file_asset_id != asset.id
+                or version.tenant_id != api_principal.tenant_id
+            ):
+                return None
+            blob = db.get(FileBlob, version.blob_id)
+            if blob is None or blob.tenant_id != api_principal.tenant_id:
+                return None
+        else:
+            try:
+                version, blob = current_version_and_blob(db, asset)
+            except FileStorageError as exc:
+                raise ManagedTabularFileUnavailableError(
+                    "Managed file version metadata is unavailable."
+                ) from exc
+        if not _is_tabular_file(version.filename_at_upload, version.content_type):
+            return None
+        return _managed_tabular_file(asset, version, blob)
+
+    def read_tabular_file(
+        self,
+        session: object,
+        principal: object,
+        *,
+        file_asset_id: str,
+        file_version_id: str,
+        max_bytes: int,
+    ) -> ManagedTabularFileContent:
+        db, api_principal, user_id = _tabular_context(
+            session,
+            principal,
+            required_scope="files:file:download",
+        )
+        try:
+            asset = get_asset_for_user(
+                db,
+                tenant_id=api_principal.tenant_id,
+                user_id=user_id,
+                asset_id=file_asset_id,
+                is_admin=has_scope(api_principal, "files:file:admin"),
+            )
+        except FileStorageError as exc:
+            raise ManagedTabularFileNotFoundError(
+                "Managed tabular file not found."
+            ) from exc
+        metadata = self.get_tabular_file(
+            db,
+            api_principal,
+            file_asset_id=asset.id,
+            file_version_id=file_version_id,
+        )
+        if metadata is None:
+            raise ManagedTabularFileNotFoundError(
+                "Managed tabular file version not found."
+            )
+        effective_max = max(1, int(max_bytes))
+        if metadata.size_bytes > effective_max:
+            raise ManagedTabularFileValidationError(
+                f"Managed tabular files are limited to {effective_max:,} bytes for this operation."
+            )
+        try:
+            payload, version, blob = read_asset_version_bytes(
+                db,
+                asset,
+                file_version_id,
+            )
+        except FileStorageError as exc:
+            raise ManagedTabularFileUnavailableError(
+                "Managed tabular file content is unavailable or failed integrity verification."
+            ) from exc
+        if len(payload) > effective_max:
+            raise ManagedTabularFileValidationError(
+                f"Managed tabular files are limited to {effective_max:,} bytes for this operation."
+            )
+        result = _managed_tabular_file(asset, version, blob)
+        audit_from_principal(
+            db,
+            api_principal,
+            action="files.tabular_content.read",
+            object_type="file_version",
+            object_id=version.id,
+            details={
+                "file_asset_id": asset.id,
+                "size_bytes": version.size_bytes,
+                "checksum_sha256": version.checksum_sha256,
+                "consumer": "tabular_content",
+            },
+        )
+        return ManagedTabularFileContent(file=result, payload=payload)
+
+
+def managed_tabular_file_capability(
+    context: ModuleContext,
+) -> FilesManagedTabularFileService:
+    configure_runtime(registry=context.registry, settings=context.settings)
+    return FilesManagedTabularFileService()
+
+
+def _tabular_context(
+    session: object,
+    principal: object,
+    *,
+    required_scope: str,
+) -> tuple[Session, ApiPrincipal, str]:
+    if not isinstance(session, Session):
+        raise TypeError("Managed tabular file access requires a SQLAlchemy session.")
+    if not isinstance(principal, ApiPrincipal):
+        raise ManagedTabularFileAccessError(
+            "Managed tabular file access requires a tenant API principal."
+        )
+    if not (
+        has_scope(principal, required_scope)
+        or has_scope(principal, "files:file:admin")
+    ):
+        raise ManagedTabularFileAccessError(
+            f"Managed tabular file access requires {required_scope}."
+        )
+    user_id = str(getattr(principal.user, "id", "") or "").strip()
+    if not principal.tenant_id or not user_id:
+        raise ManagedTabularFileAccessError(
+            "Managed tabular file access requires a tenant user principal."
+        )
+    return session, principal, user_id
+
+
+def _managed_tabular_file(
+    asset: FileAsset,
+    version: FileVersion,
+    blob: FileBlob,
+) -> ManagedTabularFile:
+    return ManagedTabularFile(
+        file_asset_id=asset.id,
+        file_version_id=version.id,
+        filename=version.filename_at_upload or asset.filename,
+        display_path=version.display_path_at_upload or asset.display_path,
+        content_type=version.content_type or blob.content_type,
+        size_bytes=version.size_bytes,
+        sha256=version.checksum_sha256,
+        updated_at=version.created_at,
+        current_version=asset.current_version_id == version.id,
+    )
+
+
+def _is_tabular_file(filename: str, content_type: str | None) -> bool:
+    normalized_name = str(filename or "").strip().casefold()
+    normalized_type = str(content_type or "").split(";", 1)[0].strip().casefold()
+    return normalized_name.endswith((".csv", ".xlsx")) or normalized_type in {
+        "text/csv",
+        "application/csv",
+        "application/vnd.ms-excel",
+        "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet",
+    }
+
+
 def virtual_folder_resource_id(*, tenant_id: str, owner_type: str, owner_id: str, path: str) -> str:
     normalized_path = normalize_folder(path)
     encoded_path = base64.urlsafe_b64encode(normalized_path.encode("utf-8")).decode("ascii").rstrip("=")

src/govoplan_files/backend/change_tracking.py

@@ -56,6 +56,10 @@ def _record_asset_change(session: OrmSession, asset: FileAsset) -> None:
             "filename",
             "description",
             "deleted_at",
+            "retained_until",
+            "legal_hold",
+            "lifecycle_revision",
+            "lifecycle_reason",
             "metadata_",
         ),
     )
@@ -79,6 +83,9 @@ def _record_asset_change(session: OrmSession, asset: FileAsset) -> None:
             "previous_path": previous_value(asset, "display_path"),
             "filename": asset.filename,
             "deleted_at": _isoformat(asset.deleted_at),
+            "retained_until": _isoformat(asset.retained_until),
+            "legal_hold": asset.legal_hold,
+            "lifecycle_revision": asset.lifecycle_revision,
         },
     )
 
@@ -117,7 +124,15 @@ def _record_share_visibility_change(session: OrmSession, share: FileShare) -> No
         return
     if not state.pending and not has_attr_changes(
         state,
-        ("file_asset_id", "target_type", "target_id", "permission", "revoked_at"),
+        (
+            "file_asset_id",
+            "target_type",
+            "target_id",
+            "permission",
+            "expires_at",
+            "revoked_at",
+            "revoked_by_user_id",
+        ),
     ):
         return
     _ensure_id(share)
@@ -136,7 +151,9 @@ def _record_share_visibility_change(session: OrmSession, share: FileShare) -> No
             "share_target_type": share.target_type,
             "share_target_id": share.target_id,
             "share_permission": share.permission,
+            "share_expires_at": _isoformat(share.expires_at),
             "share_revoked_at": _isoformat(share.revoked_at),
+            "share_revoked_by_user_id": share.revoked_by_user_id,
         },
     )
 

src/govoplan_files/backend/configuration_provider.py

@@ -0,0 +1,600 @@
+from __future__ import annotations
+
+from collections.abc import Mapping
+import os
+from pathlib import Path
+from typing import Any
+from urllib.parse import urlsplit
+
+from govoplan_core.core.configuration_packages import (
+    ConfigurationApplyResult,
+    ConfigurationDiagnostic,
+    ConfigurationExportResult,
+    ConfigurationExportSelection,
+    ConfigurationPackageFragment,
+    ConfigurationPlanItem,
+    ConfigurationPreflightContext,
+    ConfigurationPreflightResult,
+    ConfigurationProvider,
+    ConfigurationProviderDescription,
+)
+from govoplan_core.core.infrastructure_capabilities import (
+    InfrastructureCapability,
+    InfrastructureCapabilityReceipt,
+)
+from govoplan_files.backend.runtime import settings as runtime_settings
+
+
+FILES_CONFIGURATION_CAPABILITY = "files.configuration"
+MANAGED_STORAGE_FRAGMENT = "managed_storage"
+_PAYLOAD_KEYS = frozenset(
+    {"capability_id", "expected_backend", "expected_source"}
+)
+
+
+class FilesConfigurationProvider(ConfigurationProvider):
+    module_id = "files"
+
+    def __init__(
+        self,
+        *,
+        settings: object | None = None,
+        environment: Mapping[str, str] | None = None,
+    ) -> None:
+        self._settings = runtime_settings if settings is None else settings
+        self._environment = os.environ if environment is None else environment
+
+    def describe(self) -> ConfigurationProviderDescription:
+        return ConfigurationProviderDescription(
+            module_id=self.module_id,
+            fragment_types=(MANAGED_STORAGE_FRAGMENT,),
+            schema_refs={
+                MANAGED_STORAGE_FRAGMENT: "govoplan/files/configuration/managed-storage.v1"
+            },
+            exported_scopes=("system",),
+        )
+
+    def preflight(
+        self,
+        fragment: ConfigurationPackageFragment,
+        context: ConfigurationPreflightContext,
+    ) -> ConfigurationPreflightResult:
+        if fragment.fragment_type != MANAGED_STORAGE_FRAGMENT:
+            return ConfigurationPreflightResult(
+                diagnostics=(_unsupported(fragment),),
+                plan=(_blocked_plan(fragment, "Fragment type is unsupported."),),
+            )
+        diagnostics, binding_ref = self._binding_diagnostics(fragment, context)
+        blocked = any(item.severity == "blocker" for item in diagnostics)
+        return ConfigurationPreflightResult(
+            diagnostics=tuple(diagnostics),
+            plan=(
+                ConfigurationPlanItem(
+                    action="blocked" if blocked else "skip",
+                    module_id="files",
+                    fragment_type=MANAGED_STORAGE_FRAGMENT,
+                    fragment_id=fragment.fragment_id or binding_ref,
+                    summary=(
+                        "Managed storage does not match the deployment receipt."
+                        if blocked
+                        else "Deployment-owned managed storage already matches the receipt; no module state is rewritten."
+                    ),
+                ),
+            ),
+        )
+
+    def apply(
+        self,
+        fragment: ConfigurationPackageFragment,
+        supplied_data: Mapping[str, Any],
+        context: ConfigurationPreflightContext,
+    ) -> ConfigurationApplyResult:
+        del supplied_data
+        if fragment.fragment_type != MANAGED_STORAGE_FRAGMENT:
+            return ConfigurationApplyResult(diagnostics=(_unsupported(fragment),))
+        diagnostics, _binding_ref = self._binding_diagnostics(fragment, context)
+        return ConfigurationApplyResult(
+            diagnostics=tuple(
+                item for item in diagnostics if item.severity == "blocker"
+            )
+        )
+
+    def export(
+        self,
+        selection: ConfigurationExportSelection,
+        context: ConfigurationPreflightContext,
+    ) -> ConfigurationExportResult:
+        del selection
+        receipt = context.infrastructure_receipt
+        if context.infrastructure_receipt_error:
+            return ConfigurationExportResult(
+                diagnostics=(
+                    _receipt_error(
+                        context.infrastructure_receipt_error,
+                        object_ref="files.storage",
+                    ),
+                )
+            )
+        if receipt is None:
+            return ConfigurationExportResult(
+                diagnostics=(
+                    ConfigurationDiagnostic(
+                        severity="blocker",
+                        code="infrastructure_receipt_missing",
+                        message="Files managed-storage export requires the deployment capability receipt.",
+                        module_id="files",
+                        object_ref="files.storage",
+                        resolution="Mount the installer-generated receipt before exporting deployment configuration.",
+                    ),
+                )
+            )
+        capability = receipt.capability("files.storage")
+        if capability is None:
+            return ConfigurationExportResult(
+                diagnostics=(
+                    ConfigurationDiagnostic(
+                        severity="blocker",
+                        code="infrastructure_capability_missing",
+                        message="The deployment receipt does not declare managed file storage.",
+                        module_id="files",
+                        object_ref="files.storage",
+                    ),
+                )
+            )
+        return ConfigurationExportResult(
+            fragments=(
+                ConfigurationPackageFragment(
+                    module_id="files",
+                    fragment_type=MANAGED_STORAGE_FRAGMENT,
+                    fragment_id=f"{receipt.installation_id}:files.storage",
+                    payload={
+                        "capability_id": "files.storage",
+                        "expected_backend": _expected_backend(capability.source),
+                        "expected_source": capability.source,
+                    },
+                ),
+            )
+        )
+
+    def health(
+        self,
+        import_result: ConfigurationApplyResult,
+        context: ConfigurationPreflightContext,
+    ) -> tuple[ConfigurationDiagnostic, ...]:
+        del context
+        return tuple(
+            item for item in import_result.diagnostics if item.severity == "blocker"
+        )
+
+    def _binding_diagnostics(
+        self,
+        fragment: ConfigurationPackageFragment,
+        context: ConfigurationPreflightContext,
+    ) -> tuple[list[ConfigurationDiagnostic], str]:
+        diagnostics: list[ConfigurationDiagnostic] = []
+        payload = _files_fragment_payload(fragment, diagnostics)
+        if payload is None:
+            return diagnostics, fragment.fragment_id or "files.storage"
+        receipt_state = _files_receipt_capability(
+            fragment,
+            context,
+            payload,
+            diagnostics,
+        )
+        if receipt_state is None:
+            return diagnostics, fragment.fragment_id or "files.storage"
+        _receipt, capability, binding_ref = receipt_state
+        expected_backend = _expected_backend(capability.source)
+        diagnostics.extend(
+            _package_expectation_diagnostics(
+                payload,
+                capability,
+                expected_backend,
+                binding_ref,
+            )
+        )
+        active_backend = _normalized_backend(
+            getattr(self._settings, "file_storage_backend", "local")
+        )
+        if active_backend != expected_backend:
+            diagnostics.append(
+                ConfigurationDiagnostic(
+                    severity="blocker",
+                    code="files_storage_runtime_mismatch",
+                    message=(
+                        f"Files runtime uses {active_backend!r}, while the deployment receipt declares {expected_backend!r}."
+                    ),
+                    module_id="files",
+                    object_ref=binding_ref,
+                    resolution="Reconcile deployment environment and receipt before starting or importing Files configuration.",
+                )
+            )
+            return diagnostics, binding_ref
+        diagnostics.extend(
+            self._runtime_storage_diagnostics(
+                expected_backend,
+                capability,
+                binding_ref,
+            )
+        )
+        if "files" not in capability.dependent_modules:
+            diagnostics.append(
+                ConfigurationDiagnostic(
+                    severity="info",
+                    code="infrastructure_consumer_not_declared",
+                    message="Files is active but was not selected as a receipt consumer when the deployment plan was generated.",
+                    module_id="files",
+                    object_ref=binding_ref,
+                    resolution="Regenerate the deployment plan so removal-impact inventory includes Files.",
+                )
+            )
+        return diagnostics, binding_ref
+
+    def _runtime_storage_diagnostics(
+        self,
+        expected_backend: str,
+        capability: InfrastructureCapability,
+        binding_ref: str,
+    ) -> list[ConfigurationDiagnostic]:
+        if expected_backend == "s3":
+            return self._s3_diagnostics(
+                capability.endpoint,
+                capability.secret_refs,
+                capability.source,
+                binding_ref,
+            )
+        root = Path(
+            str(getattr(self._settings, "file_storage_local_root", "") or "")
+        )
+        if root.is_absolute():
+            return []
+        return [
+            ConfigurationDiagnostic(
+                severity="blocker",
+                code="files_storage_local_root_not_durable",
+                message="Receipt-bound local file storage requires an absolute deployment-managed path.",
+                module_id="files",
+                object_ref=binding_ref,
+                resolution="Set FILE_STORAGE_LOCAL_ROOT to the mounted persistent-volume path.",
+            )
+        ]
+
+    def _s3_diagnostics(
+        self,
+        endpoint: Mapping[str, object],
+        secret_refs: tuple[str, ...],
+        source: str,
+        object_ref: str,
+    ) -> list[ConfigurationDiagnostic]:
+        diagnostics: list[ConfigurationDiagnostic] = []
+        endpoint_url = str(
+            getattr(self._settings, "file_storage_s3_endpoint_url", "")
+            or getattr(self._settings, "s3_endpoint_url", "")
+            or ""
+        ).strip()
+        runtime_endpoint = _redacted_endpoint(
+            endpoint_url,
+            default_port=int(endpoint.get("port") or 443),
+        )
+        receipt_endpoint = {
+            key: endpoint.get(key) for key in ("scheme", "host", "port") if endpoint.get(key) is not None
+        }
+        if receipt_endpoint and runtime_endpoint != receipt_endpoint:
+            diagnostics.append(
+                ConfigurationDiagnostic(
+                    severity="blocker",
+                    code="files_storage_endpoint_mismatch",
+                    message="Files S3 endpoint does not match the sanitized deployment receipt endpoint.",
+                    module_id="files",
+                    object_ref=object_ref,
+                    resolution="Reconcile FILE_STORAGE_S3_ENDPOINT_URL and the deployment plan before apply.",
+                )
+            )
+        bucket = str(
+            getattr(self._settings, "file_storage_s3_bucket", "")
+            or getattr(self._settings, "s3_bucket", "")
+            or ""
+        ).strip()
+        if not bucket:
+            diagnostics.append(
+                ConfigurationDiagnostic(
+                    severity="blocker",
+                    code="files_storage_bucket_missing",
+                    message="Files S3 storage has no configured bucket.",
+                    module_id="files",
+                    object_ref=object_ref,
+                    resolution="Set FILE_STORAGE_S3_BUCKET in the deployment environment.",
+                )
+            )
+        for reference in secret_refs:
+            variable = reference.removeprefix("env:")
+            if not variable.startswith("FILE_STORAGE_"):
+                continue
+            if not str(self._environment.get(variable, "")).strip():
+                diagnostics.append(
+                    ConfigurationDiagnostic(
+                        severity="blocker",
+                        code="files_storage_secret_reference_unresolved",
+                        message=f"Required storage secret reference {reference} is not available to the runtime.",
+                        module_id="files",
+                        object_ref=object_ref,
+                        resolution="Provide the referenced environment secret without copying its value into the package.",
+                    )
+                )
+        deployment_managed = bool(
+            getattr(
+                self._settings,
+                "file_storage_s3_deployment_managed",
+                False,
+            )
+        )
+        endpoint_trusted = bool(
+            getattr(self._settings, "file_storage_s3_endpoint_trusted", False)
+        )
+        if source == "installer-managed-garage" and not deployment_managed:
+            diagnostics.append(
+                ConfigurationDiagnostic(
+                    severity="blocker",
+                    code="files_storage_management_boundary_mismatch",
+                    message="Garage is installer-managed in the receipt but not marked deployment-managed in Files runtime.",
+                    module_id="files",
+                    object_ref=object_ref,
+                    resolution="Set FILE_STORAGE_S3_DEPLOYMENT_MANAGED=true for the installer-managed Garage endpoint.",
+                )
+            )
+        if source == "operator-supplied-s3" and not endpoint_trusted:
+            diagnostics.append(
+                ConfigurationDiagnostic(
+                    severity="blocker",
+                    code="files_storage_trust_boundary_missing",
+                    message="External S3 storage must be explicitly marked as a trusted deployment endpoint.",
+                    module_id="files",
+                    object_ref=object_ref,
+                    resolution="Review the endpoint and set FILE_STORAGE_S3_ENDPOINT_TRUSTED=true in deployment configuration.",
+                )
+            )
+        return diagnostics
+
+
+def _files_fragment_payload(
+    fragment: ConfigurationPackageFragment,
+    diagnostics: list[ConfigurationDiagnostic],
+) -> Mapping[str, Any] | None:
+    payload = fragment.payload
+    if not isinstance(payload, Mapping):
+        diagnostics.append(
+            _invalid(fragment, "Managed-storage payload must be an object.")
+        )
+        return None
+    unknown = sorted(set(payload) - _PAYLOAD_KEYS)
+    if not unknown:
+        return payload
+    secret_like = any(
+        marker in key.casefold()
+        for key in unknown
+        for marker in ("password", "secret", "token", "credential", "access_key")
+    )
+    diagnostics.append(
+        ConfigurationDiagnostic(
+            severity="blocker",
+            code=(
+                "files_configuration_secret_forbidden"
+                if secret_like
+                else "files_configuration_payload_invalid"
+            ),
+            message=(
+                "Managed-storage fragments accept receipt references and non-secret expectations only."
+                if secret_like
+                else f"Managed-storage payload contains unsupported fields: {', '.join(unknown)}."
+            ),
+            module_id="files",
+            object_ref=fragment.fragment_id or MANAGED_STORAGE_FRAGMENT,
+            resolution="Keep storage credentials in deployment environment references, never in a configuration package.",
+        )
+    )
+    return payload
+
+
+def _files_receipt_capability(
+    fragment: ConfigurationPackageFragment,
+    context: ConfigurationPreflightContext,
+    payload: Mapping[str, Any],
+    diagnostics: list[ConfigurationDiagnostic],
+) -> tuple[
+    InfrastructureCapabilityReceipt,
+    InfrastructureCapability,
+    str,
+] | None:
+    if context.infrastructure_receipt_error:
+        diagnostics.append(
+            _receipt_error(
+                context.infrastructure_receipt_error,
+                object_ref=fragment.fragment_id or "files.storage",
+            )
+        )
+        return None
+    receipt = context.infrastructure_receipt
+    if receipt is None:
+        diagnostics.append(
+            ConfigurationDiagnostic(
+                severity="blocker",
+                code="infrastructure_receipt_missing",
+                message="Files managed-storage configuration requires the deployment capability receipt.",
+                module_id="files",
+                object_ref=fragment.fragment_id or "files.storage",
+                resolution="Mount the installer-generated receipt and rerun preflight.",
+            )
+        )
+        return None
+    binding_ref = fragment.fragment_id or f"{receipt.installation_id}:files.storage"
+    capability_id = _text(payload.get("capability_id")) or "files.storage"
+    if capability_id != "files.storage":
+        diagnostics.append(
+            _invalid(
+                fragment,
+                "Files managed-storage fragments must reference capability 'files.storage'.",
+            )
+        )
+        return None
+    capability = receipt.capability(capability_id)
+    if capability is None:
+        diagnostics.append(
+            ConfigurationDiagnostic(
+                severity="blocker",
+                code="infrastructure_capability_missing",
+                message="The deployment receipt does not declare managed file storage.",
+                module_id="files",
+                object_ref=binding_ref,
+                resolution="Regenerate the receipt from a deployment profile that declares managed storage.",
+            )
+        )
+        return None
+    state_diagnostic = _storage_capability_state_diagnostic(capability, binding_ref)
+    if state_diagnostic is not None:
+        diagnostics.append(state_diagnostic)
+        return None
+    return receipt, capability, binding_ref
+
+
+def _storage_capability_state_diagnostic(
+    capability: InfrastructureCapability,
+    binding_ref: str,
+) -> ConfigurationDiagnostic | None:
+    if capability.state == "unavailable":
+        return ConfigurationDiagnostic(
+            severity="blocker",
+            code="infrastructure_capability_unavailable",
+            message="The deployment receipt states that managed file storage is unavailable.",
+            module_id="files",
+            object_ref=binding_ref,
+            resolution="Select local, managed Garage, or external S3 storage in the deployment profile.",
+        )
+    if capability.state == "available_unconfigured":
+        return ConfigurationDiagnostic(
+            severity="blocker",
+            code="infrastructure_capability_unconfigured",
+            message="Managed file storage is available but has not been bound by the deployment runtime.",
+            module_id="files",
+            object_ref=binding_ref,
+            resolution="Complete the deployment-owned storage configuration before importing Files configuration.",
+        )
+    return None
+
+
+def _package_expectation_diagnostics(
+    payload: Mapping[str, Any],
+    capability: InfrastructureCapability,
+    expected_backend: str,
+    binding_ref: str,
+) -> list[ConfigurationDiagnostic]:
+    diagnostics: list[ConfigurationDiagnostic] = []
+    package_backend = (
+        _text(payload.get("expected_backend")) or expected_backend
+    ).casefold()
+    if package_backend != expected_backend:
+        diagnostics.append(
+            ConfigurationDiagnostic(
+                severity="blocker",
+                code="infrastructure_backend_mismatch",
+                message="The package storage backend expectation conflicts with the deployment receipt.",
+                module_id="files",
+                object_ref=binding_ref,
+                resolution="Use the receipt backend or regenerate the deployment package after review.",
+            )
+        )
+    package_source = _text(payload.get("expected_source"))
+    if package_source and package_source != capability.source:
+        diagnostics.append(
+            ConfigurationDiagnostic(
+                severity="blocker",
+                code="infrastructure_source_mismatch",
+                message="The package storage source expectation conflicts with the deployment receipt.",
+                module_id="files",
+                object_ref=binding_ref,
+                resolution="Review the provider replacement and regenerate the package from the active receipt.",
+            )
+        )
+    return diagnostics
+
+
+def _expected_backend(source: str) -> str:
+    return "local" if source == "host-local" else "s3"
+
+
+def _normalized_backend(value: object) -> str:
+    clean = str(value or "local").strip().casefold()
+    if clean in {"local", "filesystem", "fs"}:
+        return "local"
+    if clean in {"s3", "garage"}:
+        return "s3"
+    return clean
+
+
+def _redacted_endpoint(value: str, *, default_port: int) -> dict[str, object]:
+    try:
+        parsed = urlsplit(value)
+        host = parsed.hostname
+        if not parsed.scheme or not host:
+            return {"reference": "unresolved"}
+        port = parsed.port or default_port
+    except ValueError:
+        return {"reference": "unresolved"}
+    return {"scheme": parsed.scheme, "host": host, "port": port}
+
+
+def _text(value: object) -> str | None:
+    if value is None:
+        return None
+    text = str(value).strip()
+    return text or None
+
+
+def _receipt_error(error: str, *, object_ref: str) -> ConfigurationDiagnostic:
+    return ConfigurationDiagnostic(
+        severity="blocker",
+        code="infrastructure_receipt_invalid",
+        message=f"The deployment capability receipt is invalid: {error}",
+        module_id="files",
+        object_ref=object_ref,
+        resolution="Repair or regenerate the deployment receipt before importing Files configuration.",
+    )
+
+
+def _invalid(
+    fragment: ConfigurationPackageFragment,
+    message: str,
+) -> ConfigurationDiagnostic:
+    return ConfigurationDiagnostic(
+        severity="blocker",
+        code="files_configuration_payload_invalid",
+        message=message,
+        module_id="files",
+        object_ref=fragment.fragment_id or fragment.fragment_type,
+        resolution="Review the Files configuration-package fragment and rerun preflight.",
+    )
+
+
+def _unsupported(fragment: ConfigurationPackageFragment) -> ConfigurationDiagnostic:
+    return ConfigurationDiagnostic(
+        severity="blocker",
+        code="fragment_type_unsupported",
+        message=f"Files configuration does not support fragment type {fragment.fragment_type!r}.",
+        module_id="files",
+        object_ref=fragment.fragment_id or fragment.fragment_type,
+    )
+
+
+def _blocked_plan(
+    fragment: ConfigurationPackageFragment,
+    summary: str,
+) -> ConfigurationPlanItem:
+    return ConfigurationPlanItem(
+        action="blocked",
+        module_id="files",
+        fragment_type=fragment.fragment_type,
+        fragment_id=fragment.fragment_id,
+        summary=summary,
+    )
+
+
+__all__ = ["FILES_CONFIGURATION_CAPABILITY", "FilesConfigurationProvider"]

src/govoplan_files/backend/db/models.py

@@ -4,7 +4,18 @@ import uuid
 from datetime import datetime
 from typing import Any
 
-from sqlalchemy import Boolean, DateTime, ForeignKey, Index, Integer, JSON, String, Text, UniqueConstraint, text
+from sqlalchemy import (
+    Boolean,
+    DateTime,
+    ForeignKey,
+    Index,
+    Integer,
+    JSON,
+    String,
+    Text,
+    UniqueConstraint,
+    text,
+)
 from sqlalchemy.orm import Mapped, mapped_column
 
 from govoplan_core.db.base import Base, TimestampMixin
@@ -16,7 +27,15 @@ def new_uuid() -> str:
 
 class FileBlob(Base, TimestampMixin):
     __tablename__ = "file_blobs"
-    __table_args__ = (UniqueConstraint("tenant_id", "checksum_sha256", "size_bytes", name="uq_file_blobs_tenant_checksum_size"),)
+    __table_args__ = (
+        UniqueConstraint(
+            "tenant_id",
+            "checksum_sha256",
+            "size_bytes",
+            "protection_discriminator",
+            name="uq_file_blobs_tenant_checksum_size_protection",
+        ),
+    )
 
     id: Mapped[str] = mapped_column(String(36), primary_key=True, default=new_uuid)
     tenant_id: Mapped[str] = mapped_column(String(36), nullable=False, index=True)
@@ -25,9 +44,106 @@ class FileBlob(Base, TimestampMixin):
     storage_key: Mapped[str] = mapped_column(String(1000), nullable=False)
     checksum_sha256: Mapped[str] = mapped_column(String(64), nullable=False, index=True)
     size_bytes: Mapped[int] = mapped_column(Integer, nullable=False)
+    protection_discriminator: Mapped[str] = mapped_column(
+        String(320), default="plaintext", nullable=False, index=True
+    )
+    encryption_envelope_id: Mapped[str | None] = mapped_column(
+        String(255), nullable=True, index=True
+    )
+    storage_checksum_sha256: Mapped[str | None] = mapped_column(
+        String(64), nullable=True
+    )
+    storage_size_bytes: Mapped[int | None] = mapped_column(Integer, nullable=True)
     content_type: Mapped[str | None] = mapped_column(String(255))
     ref_count: Mapped[int] = mapped_column(Integer, default=1, nullable=False)
     retained_until: Mapped[datetime | None] = mapped_column(DateTime(timezone=True))
+    integrity_status: Mapped[str] = mapped_column(
+        String(30), default="unchecked", nullable=False, index=True
+    )
+    integrity_checked_at: Mapped[datetime | None] = mapped_column(
+        DateTime(timezone=True), nullable=True, index=True
+    )
+    integrity_failure: Mapped[str | None] = mapped_column(String(100), nullable=True)
+    quarantined_at: Mapped[datetime | None] = mapped_column(
+        DateTime(timezone=True), nullable=True, index=True
+    )
+
+
+class FileIntegrityScan(Base, TimestampMixin):
+    __tablename__ = "file_integrity_scans"
+
+    id: Mapped[str] = mapped_column(String(36), primary_key=True, default=new_uuid)
+    tenant_id: Mapped[str] = mapped_column(String(36), nullable=False, index=True)
+    storage_backend: Mapped[str] = mapped_column(String(50), nullable=False)
+    storage_prefix: Mapped[str] = mapped_column(String(1000), nullable=False)
+    status: Mapped[str] = mapped_column(
+        String(30), default="pending", nullable=False, index=True
+    )
+    revision: Mapped[int] = mapped_column(Integer, default=1, nullable=False)
+    phase: Mapped[str] = mapped_column(String(30), default="blobs", nullable=False)
+    verify_checksums: Mapped[bool] = mapped_column(
+        Boolean, default=True, nullable=False
+    )
+    batch_size: Mapped[int] = mapped_column(Integer, default=100, nullable=False)
+    blob_cursor: Mapped[str | None] = mapped_column(String(36), nullable=True)
+    object_cursor: Mapped[str | None] = mapped_column(String(1000), nullable=True)
+    scanned_blob_count: Mapped[int] = mapped_column(Integer, default=0, nullable=False)
+    verified_blob_count: Mapped[int] = mapped_column(Integer, default=0, nullable=False)
+    quarantined_blob_count: Mapped[int] = mapped_column(
+        Integer, default=0, nullable=False
+    )
+    scanned_object_count: Mapped[int] = mapped_column(
+        Integer, default=0, nullable=False
+    )
+    orphan_object_count: Mapped[int] = mapped_column(Integer, default=0, nullable=False)
+    created_by_user_id: Mapped[str | None] = mapped_column(
+        ForeignKey("access_users.id", ondelete="SET NULL"), nullable=True, index=True
+    )
+    started_at: Mapped[datetime | None] = mapped_column(
+        DateTime(timezone=True), nullable=True
+    )
+    completed_at: Mapped[datetime | None] = mapped_column(
+        DateTime(timezone=True), nullable=True
+    )
+    last_error: Mapped[str | None] = mapped_column(String(255), nullable=True)
+
+
+class FileIntegrityFinding(Base, TimestampMixin):
+    __tablename__ = "file_integrity_findings"
+    __table_args__ = (
+        Index("ix_file_integrity_findings_scan_state", "scan_id", "state"),
+    )
+
+    id: Mapped[str] = mapped_column(String(36), primary_key=True, default=new_uuid)
+    scan_id: Mapped[str] = mapped_column(
+        ForeignKey("file_integrity_scans.id", ondelete="CASCADE"),
+        nullable=False,
+        index=True,
+    )
+    tenant_id: Mapped[str] = mapped_column(String(36), nullable=False, index=True)
+    kind: Mapped[str] = mapped_column(String(40), nullable=False, index=True)
+    state: Mapped[str] = mapped_column(
+        String(30), default="open", nullable=False, index=True
+    )
+    revision: Mapped[int] = mapped_column(Integer, default=1, nullable=False)
+    blob_id: Mapped[str | None] = mapped_column(
+        ForeignKey("file_blobs.id", ondelete="SET NULL"), nullable=True, index=True
+    )
+    storage_key: Mapped[str] = mapped_column(String(1000), nullable=False)
+    expected_size_bytes: Mapped[int | None] = mapped_column(Integer, nullable=True)
+    observed_size_bytes: Mapped[int | None] = mapped_column(Integer, nullable=True)
+    expected_checksum_sha256: Mapped[str | None] = mapped_column(
+        String(64), nullable=True
+    )
+    observed_checksum_sha256: Mapped[str | None] = mapped_column(
+        String(64), nullable=True
+    )
+    resolved_at: Mapped[datetime | None] = mapped_column(
+        DateTime(timezone=True), nullable=True
+    )
+    resolved_by_user_id: Mapped[str | None] = mapped_column(
+        ForeignKey("access_users.id", ondelete="SET NULL"), nullable=True, index=True
+    )
 
 
 class FileFolder(Base, TimestampMixin):
@@ -35,14 +151,18 @@ class FileFolder(Base, TimestampMixin):
     __table_args__ = (
         Index(
             "uq_file_folders_active_user_path",
-            "tenant_id", "owner_user_id", "path",
+            "tenant_id",
+            "owner_user_id",
+            "path",
             unique=True,
             sqlite_where=text("owner_type = 'user' AND deleted_at IS NULL"),
             postgresql_where=text("owner_type = 'user' AND deleted_at IS NULL"),
         ),
         Index(
             "uq_file_folders_active_group_path",
-            "tenant_id", "owner_group_id", "path",
+            "tenant_id",
+            "owner_group_id",
+            "path",
             unique=True,
             sqlite_where=text("owner_type = 'group' AND deleted_at IS NULL"),
             postgresql_where=text("owner_type = 'group' AND deleted_at IS NULL"),
@@ -52,12 +172,22 @@ class FileFolder(Base, TimestampMixin):
     id: Mapped[str] = mapped_column(String(36), primary_key=True, default=new_uuid)
     tenant_id: Mapped[str] = mapped_column(String(36), nullable=False, index=True)
     owner_type: Mapped[str] = mapped_column(String(20), nullable=False, index=True)
-    owner_user_id: Mapped[str | None] = mapped_column(ForeignKey("access_users.id", ondelete="SET NULL"), nullable=True, index=True)
+    owner_user_id: Mapped[str | None] = mapped_column(
-    owner_group_id: Mapped[str | None] = mapped_column(ForeignKey("access_groups.id", ondelete="SET NULL"), nullable=True, index=True)
+        ForeignKey("access_users.id", ondelete="SET NULL"), nullable=True, index=True
+    )
+    owner_group_id: Mapped[str | None] = mapped_column(
+        ForeignKey("access_groups.id", ondelete="SET NULL"), nullable=True, index=True
+    )
     path: Mapped[str] = mapped_column(String(1000), nullable=False, index=True)
-    created_by_user_id: Mapped[str | None] = mapped_column(ForeignKey("access_users.id", ondelete="SET NULL"), nullable=True, index=True)
+    created_by_user_id: Mapped[str | None] = mapped_column(
-    deleted_at: Mapped[datetime | None] = mapped_column(DateTime(timezone=True), nullable=True, index=True)
+        ForeignKey("access_users.id", ondelete="SET NULL"), nullable=True, index=True
-    metadata_: Mapped[dict[str, Any] | None] = mapped_column("metadata", JSON, nullable=True)
+    )
+    deleted_at: Mapped[datetime | None] = mapped_column(
+        DateTime(timezone=True), nullable=True, index=True
+    )
+    metadata_: Mapped[dict[str, Any] | None] = mapped_column(
+        "metadata", JSON, nullable=True
+    )
 
 
 class FileAsset(Base, TimestampMixin):
@@ -66,46 +196,159 @@ class FileAsset(Base, TimestampMixin):
     id: Mapped[str] = mapped_column(String(36), primary_key=True, default=new_uuid)
     tenant_id: Mapped[str] = mapped_column(String(36), nullable=False, index=True)
     owner_type: Mapped[str] = mapped_column(String(20), nullable=False, index=True)
-    owner_user_id: Mapped[str | None] = mapped_column(ForeignKey("access_users.id", ondelete="SET NULL"), nullable=True, index=True)
+    owner_user_id: Mapped[str | None] = mapped_column(
-    owner_group_id: Mapped[str | None] = mapped_column(ForeignKey("access_groups.id", ondelete="SET NULL"), nullable=True, index=True)
+        ForeignKey("access_users.id", ondelete="SET NULL"), nullable=True, index=True
-    current_version_id: Mapped[str | None] = mapped_column(String(36), nullable=True, index=True)
+    )
+    owner_group_id: Mapped[str | None] = mapped_column(
+        ForeignKey("access_groups.id", ondelete="SET NULL"), nullable=True, index=True
+    )
+    current_version_id: Mapped[str | None] = mapped_column(
+        String(36), nullable=True, index=True
+    )
     display_path: Mapped[str] = mapped_column(String(1000), nullable=False, index=True)
     filename: Mapped[str] = mapped_column(String(500), nullable=False, index=True)
     description: Mapped[str | None] = mapped_column(Text)
-    created_by_user_id: Mapped[str | None] = mapped_column(ForeignKey("access_users.id", ondelete="SET NULL"), nullable=True, index=True)
+    created_by_user_id: Mapped[str | None] = mapped_column(
-    deleted_at: Mapped[datetime | None] = mapped_column(DateTime(timezone=True), nullable=True, index=True)
+        ForeignKey("access_users.id", ondelete="SET NULL"), nullable=True, index=True
-    metadata_: Mapped[dict[str, Any] | None] = mapped_column("metadata", JSON, nullable=True)
+    )
+    deleted_at: Mapped[datetime | None] = mapped_column(
+        DateTime(timezone=True), nullable=True, index=True
+    )
+    retained_until: Mapped[datetime | None] = mapped_column(
+        DateTime(timezone=True), nullable=True, index=True
+    )
+    legal_hold: Mapped[bool] = mapped_column(
+        Boolean, default=False, nullable=False, index=True
+    )
+    lifecycle_revision: Mapped[int] = mapped_column(
+        Integer, default=1, nullable=False
+    )
+    lifecycle_reason: Mapped[str | None] = mapped_column(String(500), nullable=True)
+    metadata_: Mapped[dict[str, Any] | None] = mapped_column(
+        "metadata", JSON, nullable=True
+    )
 
 
 class FileVersion(Base, TimestampMixin):
     __tablename__ = "file_versions"
-    __table_args__ = (UniqueConstraint("file_asset_id", "version_number", name="uq_file_versions_asset_number"),)
+    __table_args__ = (
+        UniqueConstraint(
+            "file_asset_id", "version_number", name="uq_file_versions_asset_number"
+        ),
+    )
 
     id: Mapped[str] = mapped_column(String(36), primary_key=True, default=new_uuid)
     tenant_id: Mapped[str] = mapped_column(String(36), nullable=False, index=True)
-    file_asset_id: Mapped[str] = mapped_column(ForeignKey("file_assets.id", ondelete="CASCADE"), nullable=False, index=True)
+    file_asset_id: Mapped[str] = mapped_column(
-    blob_id: Mapped[str] = mapped_column(ForeignKey("file_blobs.id", ondelete="RESTRICT"), nullable=False, index=True)
+        ForeignKey("file_assets.id", ondelete="CASCADE"), nullable=False, index=True
+    )
+    blob_id: Mapped[str] = mapped_column(
+        ForeignKey("file_blobs.id", ondelete="RESTRICT"), nullable=False, index=True
+    )
     version_number: Mapped[int] = mapped_column(Integer, nullable=False)
     filename_at_upload: Mapped[str] = mapped_column(String(500), nullable=False)
     display_path_at_upload: Mapped[str] = mapped_column(String(1000), nullable=False)
     content_type: Mapped[str | None] = mapped_column(String(255))
     size_bytes: Mapped[int] = mapped_column(Integer, nullable=False)
     checksum_sha256: Mapped[str] = mapped_column(String(64), nullable=False, index=True)
-    created_by_user_id: Mapped[str | None] = mapped_column(ForeignKey("access_users.id", ondelete="SET NULL"), nullable=True, index=True)
+    created_by_user_id: Mapped[str | None] = mapped_column(
+        ForeignKey("access_users.id", ondelete="SET NULL"), nullable=True, index=True
+    )
+
+
+class FileFormEvidenceGrant(Base, TimestampMixin):
+    __tablename__ = "file_form_evidence_grants"
+    __table_args__ = (
+        UniqueConstraint(
+            "tenant_id",
+            "idempotency_key",
+            name="uq_file_form_evidence_grants_idempotency",
+        ),
+        Index(
+            "ix_file_form_evidence_grants_form",
+            "tenant_id",
+            "form_instance_id",
+            "form_definition_id",
+            "form_definition_revision",
+        ),
+    )
+
+    id: Mapped[str] = mapped_column(String(36), primary_key=True, default=new_uuid)
+    tenant_id: Mapped[str] = mapped_column(String(36), nullable=False, index=True)
+    form_instance_id: Mapped[str] = mapped_column(
+        String(36), nullable=False, index=True
+    )
+    form_definition_id: Mapped[str] = mapped_column(String(255), nullable=False)
+    form_definition_revision: Mapped[str] = mapped_column(String(255), nullable=False)
+    token_sha256: Mapped[str] = mapped_column(String(64), nullable=False, unique=True)
+    idempotency_key: Mapped[str] = mapped_column(String(255), nullable=False)
+    request_sha256: Mapped[str] = mapped_column(String(64), nullable=False)
+    custodian_user_id: Mapped[str] = mapped_column(
+        ForeignKey("access_users.id", ondelete="RESTRICT"),
+        nullable=False,
+        index=True,
+    )
+    evidence_kind: Mapped[str] = mapped_column(String(30), nullable=False)
+    purpose: Mapped[str] = mapped_column(String(500), nullable=False)
+    status: Mapped[str] = mapped_column(
+        String(30), default="issued", nullable=False, index=True
+    )
+    expires_at: Mapped[datetime] = mapped_column(
+        DateTime(timezone=True), nullable=False, index=True
+    )
+    max_size_bytes: Mapped[int] = mapped_column(Integer, nullable=False)
+    allowed_content_types: Mapped[list[str]] = mapped_column(
+        JSON, default=list, nullable=False
+    )
+    file_asset_id: Mapped[str | None] = mapped_column(
+        ForeignKey("file_assets.id", ondelete="RESTRICT"), nullable=True, index=True
+    )
+    file_version_id: Mapped[str | None] = mapped_column(
+        ForeignKey("file_versions.id", ondelete="RESTRICT"), nullable=True, index=True
+    )
+    uploaded_at: Mapped[datetime | None] = mapped_column(
+        DateTime(timezone=True), nullable=True
+    )
+    revoked_at: Mapped[datetime | None] = mapped_column(
+        DateTime(timezone=True), nullable=True
+    )
+    metadata_: Mapped[dict[str, Any]] = mapped_column(
+        "metadata", JSON, default=dict, nullable=False
+    )
 
 
 class FileShare(Base, TimestampMixin):
     __tablename__ = "file_shares"
-    __table_args__ = (UniqueConstraint("file_asset_id", "target_type", "target_id", "revoked_at", name="uq_file_shares_active_target"),)
+    __table_args__ = (
+        UniqueConstraint(
+            "file_asset_id",
+            "target_type",
+            "target_id",
+            "revoked_at",
+            name="uq_file_shares_active_target",
+        ),
+    )
 
     id: Mapped[str] = mapped_column(String(36), primary_key=True, default=new_uuid)
     tenant_id: Mapped[str] = mapped_column(String(36), nullable=False, index=True)
-    file_asset_id: Mapped[str] = mapped_column(ForeignKey("file_assets.id", ondelete="CASCADE"), nullable=False, index=True)
+    file_asset_id: Mapped[str] = mapped_column(
+        ForeignKey("file_assets.id", ondelete="CASCADE"), nullable=False, index=True
+    )
     target_type: Mapped[str] = mapped_column(String(20), nullable=False, index=True)
     target_id: Mapped[str] = mapped_column(String(36), nullable=False, index=True)
     permission: Mapped[str] = mapped_column(String(20), default="read", nullable=False)
-    created_by_user_id: Mapped[str | None] = mapped_column(ForeignKey("access_users.id", ondelete="SET NULL"), nullable=True, index=True)
+    created_by_user_id: Mapped[str | None] = mapped_column(
-    revoked_at: Mapped[datetime | None] = mapped_column(DateTime(timezone=True), nullable=True, index=True)
+        ForeignKey("access_users.id", ondelete="SET NULL"), nullable=True, index=True
+    )
+    expires_at: Mapped[datetime | None] = mapped_column(
+        DateTime(timezone=True), nullable=True, index=True
+    )
+    revoked_at: Mapped[datetime | None] = mapped_column(
+        DateTime(timezone=True), nullable=True, index=True
+    )
+    revoked_by_user_id: Mapped[str | None] = mapped_column(
+        ForeignKey("access_users.id", ondelete="SET NULL"), nullable=True, index=True
+    )
 
 
 class FileConnectorProfile(Base, TimestampMixin):
@@ -116,15 +359,23 @@ class FileConnectorProfile(Base, TimestampMixin):
 
     id: Mapped[str] = mapped_column(String(255), primary_key=True)
     tenant_id: Mapped[str | None] = mapped_column(String(36), nullable=True, index=True)
-    scope_type: Mapped[str] = mapped_column(String(20), default="tenant", nullable=False, index=True)
+    scope_type: Mapped[str] = mapped_column(
+        String(20), default="tenant", nullable=False, index=True
+    )
     scope_id: Mapped[str | None] = mapped_column(String(36), nullable=True, index=True)
     label: Mapped[str] = mapped_column(String(255), nullable=False)
     provider: Mapped[str] = mapped_column(String(50), nullable=False, index=True)
     endpoint_url: Mapped[str | None] = mapped_column(String(1000), nullable=True)
     base_path: Mapped[str | None] = mapped_column(String(1000), nullable=True)
-    enabled: Mapped[bool] = mapped_column(Boolean, default=True, nullable=False, index=True)
+    enabled: Mapped[bool] = mapped_column(
-    credential_profile_id: Mapped[str | None] = mapped_column(String(255), nullable=True, index=True)
+        Boolean, default=True, nullable=False, index=True
-    credential_mode: Mapped[str] = mapped_column(String(30), default="none", nullable=False)
+    )
+    credential_profile_id: Mapped[str | None] = mapped_column(
+        String(255), nullable=True, index=True
+    )
+    credential_mode: Mapped[str] = mapped_column(
+        String(30), default="none", nullable=False
+    )
     username: Mapped[str | None] = mapped_column(String(320), nullable=True)
     password_encrypted: Mapped[str | None] = mapped_column(Text, nullable=True)
     token_encrypted: Mapped[str | None] = mapped_column(Text, nullable=True)
@@ -133,9 +384,15 @@ class FileConnectorProfile(Base, TimestampMixin):
     secret_ref: Mapped[str | None] = mapped_column(String(1000), nullable=True)
     capabilities: Mapped[list[str] | None] = mapped_column(JSON, nullable=True)
     policy: Mapped[dict[str, Any] | None] = mapped_column(JSON, nullable=True)
-    metadata_: Mapped[dict[str, Any] | None] = mapped_column("metadata", JSON, nullable=True)
+    metadata_: Mapped[dict[str, Any] | None] = mapped_column(
-    created_by_user_id: Mapped[str | None] = mapped_column(ForeignKey("access_users.id", ondelete="SET NULL"), nullable=True, index=True)
+        "metadata", JSON, nullable=True
-    updated_by_user_id: Mapped[str | None] = mapped_column(ForeignKey("access_users.id", ondelete="SET NULL"), nullable=True, index=True)
+    )
+    created_by_user_id: Mapped[str | None] = mapped_column(
+        ForeignKey("access_users.id", ondelete="SET NULL"), nullable=True, index=True
+    )
+    updated_by_user_id: Mapped[str | None] = mapped_column(
+        ForeignKey("access_users.id", ondelete="SET NULL"), nullable=True, index=True
+    )
 
 
 class FileConnectorCredential(Base, TimestampMixin):
@@ -146,12 +403,18 @@ class FileConnectorCredential(Base, TimestampMixin):
 
     id: Mapped[str] = mapped_column(String(255), primary_key=True)
     tenant_id: Mapped[str | None] = mapped_column(String(36), nullable=True, index=True)
-    scope_type: Mapped[str] = mapped_column(String(20), default="tenant", nullable=False, index=True)
+    scope_type: Mapped[str] = mapped_column(
+        String(20), default="tenant", nullable=False, index=True
+    )
     scope_id: Mapped[str | None] = mapped_column(String(36), nullable=True, index=True)
     label: Mapped[str] = mapped_column(String(255), nullable=False)
     provider: Mapped[str | None] = mapped_column(String(50), nullable=True, index=True)
-    enabled: Mapped[bool] = mapped_column(Boolean, default=True, nullable=False, index=True)
+    enabled: Mapped[bool] = mapped_column(
-    credential_mode: Mapped[str] = mapped_column(String(30), default="none", nullable=False)
+        Boolean, default=True, nullable=False, index=True
+    )
+    credential_mode: Mapped[str] = mapped_column(
+        String(30), default="none", nullable=False
+    )
     username: Mapped[str | None] = mapped_column(String(320), nullable=True)
     password_encrypted: Mapped[str | None] = mapped_column(Text, nullable=True)
     token_encrypted: Mapped[str | None] = mapped_column(Text, nullable=True)
@@ -159,15 +422,26 @@ class FileConnectorCredential(Base, TimestampMixin):
     token_env: Mapped[str | None] = mapped_column(String(255), nullable=True)
     secret_ref: Mapped[str | None] = mapped_column(String(1000), nullable=True)
     policy: Mapped[dict[str, Any] | None] = mapped_column(JSON, nullable=True)
-    metadata_: Mapped[dict[str, Any] | None] = mapped_column("metadata", JSON, nullable=True)
+    metadata_: Mapped[dict[str, Any] | None] = mapped_column(
-    created_by_user_id: Mapped[str | None] = mapped_column(ForeignKey("access_users.id", ondelete="SET NULL"), nullable=True, index=True)
+        "metadata", JSON, nullable=True
-    updated_by_user_id: Mapped[str | None] = mapped_column(ForeignKey("access_users.id", ondelete="SET NULL"), nullable=True, index=True)
+    )
+    created_by_user_id: Mapped[str | None] = mapped_column(
+        ForeignKey("access_users.id", ondelete="SET NULL"), nullable=True, index=True
+    )
+    updated_by_user_id: Mapped[str | None] = mapped_column(
+        ForeignKey("access_users.id", ondelete="SET NULL"), nullable=True, index=True
+    )
 
 
 class FileConnectorPolicy(Base, TimestampMixin):
     __tablename__ = "file_connector_policies"
     __table_args__ = (
-        UniqueConstraint("tenant_id", "scope_type", "scope_id", name="uq_file_connector_policies_scope"),
+        UniqueConstraint(
+            "tenant_id",
+            "scope_type",
+            "scope_id",
+            name="uq_file_connector_policies_scope",
+        ),
         Index("ix_file_connector_policies_scope", "scope_type", "scope_id"),
     )
 
@@ -176,24 +450,38 @@ class FileConnectorPolicy(Base, TimestampMixin):
     scope_type: Mapped[str] = mapped_column(String(20), nullable=False, index=True)
     scope_id: Mapped[str | None] = mapped_column(String(36), nullable=True, index=True)
     policy: Mapped[dict[str, Any]] = mapped_column(JSON, default=dict, nullable=False)
-    created_by_user_id: Mapped[str | None] = mapped_column(ForeignKey("access_users.id", ondelete="SET NULL"), nullable=True, index=True)
+    created_by_user_id: Mapped[str | None] = mapped_column(
-    updated_by_user_id: Mapped[str | None] = mapped_column(ForeignKey("access_users.id", ondelete="SET NULL"), nullable=True, index=True)
+        ForeignKey("access_users.id", ondelete="SET NULL"), nullable=True, index=True
+    )
+    updated_by_user_id: Mapped[str | None] = mapped_column(
+        ForeignKey("access_users.id", ondelete="SET NULL"), nullable=True, index=True
+    )
 
 
 class FileConnectorSpace(Base, TimestampMixin):
     __tablename__ = "file_connector_spaces"
     __table_args__ = (
-        Index("ix_file_connector_spaces_owner", "tenant_id", "owner_type", "owner_user_id", "owner_group_id"),
+        Index(
+            "ix_file_connector_spaces_owner",
+            "tenant_id",
+            "owner_type",
+            "owner_user_id",
+            "owner_group_id",
+        ),
         Index(
             "uq_file_connector_spaces_active_user_label",
-            "tenant_id", "owner_user_id", "label",
+            "tenant_id",
+            "owner_user_id",
+            "label",
             unique=True,
             sqlite_where=text("owner_type = 'user' AND deleted_at IS NULL"),
             postgresql_where=text("owner_type = 'user' AND deleted_at IS NULL"),
         ),
         Index(
             "uq_file_connector_spaces_active_group_label",
-            "tenant_id", "owner_group_id", "label",
+            "tenant_id",
+            "owner_group_id",
+            "label",
             unique=True,
             sqlite_where=text("owner_type = 'group' AND deleted_at IS NULL"),
             postgresql_where=text("owner_type = 'group' AND deleted_at IS NULL"),
@@ -203,41 +491,81 @@ class FileConnectorSpace(Base, TimestampMixin):
     id: Mapped[str] = mapped_column(String(36), primary_key=True, default=new_uuid)
     tenant_id: Mapped[str] = mapped_column(String(36), nullable=False, index=True)
     owner_type: Mapped[str] = mapped_column(String(20), nullable=False, index=True)
-    owner_user_id: Mapped[str | None] = mapped_column(ForeignKey("access_users.id", ondelete="SET NULL"), nullable=True, index=True)
+    owner_user_id: Mapped[str | None] = mapped_column(
-    owner_group_id: Mapped[str | None] = mapped_column(ForeignKey("access_groups.id", ondelete="SET NULL"), nullable=True, index=True)
+        ForeignKey("access_users.id", ondelete="SET NULL"), nullable=True, index=True
+    )
+    owner_group_id: Mapped[str | None] = mapped_column(
+        ForeignKey("access_groups.id", ondelete="SET NULL"), nullable=True, index=True
+    )
     label: Mapped[str] = mapped_column(String(255), nullable=False)
-    connector_profile_id: Mapped[str] = mapped_column(String(255), nullable=False, index=True)
+    connector_profile_id: Mapped[str] = mapped_column(
+        String(255), nullable=False, index=True
+    )
     provider: Mapped[str] = mapped_column(String(50), nullable=False, index=True)
     library_id: Mapped[str | None] = mapped_column(String(255), nullable=True)
     remote_path: Mapped[str] = mapped_column(String(1000), default="", nullable=False)
     sync_mode: Mapped[str] = mapped_column(String(30), default="manual", nullable=False)
     read_only: Mapped[bool] = mapped_column(Boolean, default=True, nullable=False)
-    is_active: Mapped[bool] = mapped_column(Boolean, default=True, nullable=False, index=True)
+    is_active: Mapped[bool] = mapped_column(
-    created_by_user_id: Mapped[str | None] = mapped_column(ForeignKey("access_users.id", ondelete="SET NULL"), nullable=True, index=True)
+        Boolean, default=True, nullable=False, index=True
-    deleted_at: Mapped[datetime | None] = mapped_column(DateTime(timezone=True), nullable=True, index=True)
+    )
-    metadata_: Mapped[dict[str, Any] | None] = mapped_column("metadata", JSON, nullable=True)
+    created_by_user_id: Mapped[str | None] = mapped_column(
+        ForeignKey("access_users.id", ondelete="SET NULL"), nullable=True, index=True
+    )
+    deleted_at: Mapped[datetime | None] = mapped_column(
+        DateTime(timezone=True), nullable=True, index=True
+    )
+    metadata_: Mapped[dict[str, Any] | None] = mapped_column(
+        "metadata", JSON, nullable=True
+    )
 
 
 class CampaignAttachmentUse(Base, TimestampMixin):
     __tablename__ = "campaign_attachment_uses"
-    __table_args__ = (UniqueConstraint("campaign_job_id", "file_version_id", "filename_used", "use_stage", name="uq_campaign_attachment_uses_job_file_stage"),)
+    __table_args__ = (
+        UniqueConstraint(
+            "campaign_job_id",
+            "file_version_id",
+            "filename_used",
+            "use_stage",
+            name="uq_campaign_attachment_uses_job_file_stage",
+        ),
+    )
 
     id: Mapped[str] = mapped_column(String(36), primary_key=True, default=new_uuid)
     tenant_id: Mapped[str] = mapped_column(String(36), nullable=False, index=True)
-    campaign_id: Mapped[str] = mapped_column(ForeignKey("campaigns.id", ondelete="CASCADE"), nullable=False, index=True)
+    campaign_id: Mapped[str] = mapped_column(
-    campaign_version_id: Mapped[str] = mapped_column(ForeignKey("campaign_versions.id", ondelete="CASCADE"), nullable=False, index=True)
+        ForeignKey("campaigns.id", ondelete="CASCADE"), nullable=False, index=True
-    campaign_job_id: Mapped[str | None] = mapped_column(ForeignKey("campaign_jobs.id", ondelete="SET NULL"), nullable=True, index=True)
+    )
+    campaign_version_id: Mapped[str] = mapped_column(
+        ForeignKey("campaign_versions.id", ondelete="CASCADE"),
+        nullable=False,
+        index=True,
+    )
+    campaign_job_id: Mapped[str | None] = mapped_column(
+        ForeignKey("campaign_jobs.id", ondelete="SET NULL"), nullable=True, index=True
+    )
     entry_index: Mapped[int | None] = mapped_column(Integer)
     entry_id: Mapped[str | None] = mapped_column(String(255), index=True)
-    file_asset_id: Mapped[str] = mapped_column(ForeignKey("file_assets.id", ondelete="RESTRICT"), nullable=False, index=True)
+    file_asset_id: Mapped[str] = mapped_column(
-    file_version_id: Mapped[str] = mapped_column(ForeignKey("file_versions.id", ondelete="RESTRICT"), nullable=False, index=True)
+        ForeignKey("file_assets.id", ondelete="RESTRICT"), nullable=False, index=True
-    file_blob_id: Mapped[str] = mapped_column(ForeignKey("file_blobs.id", ondelete="RESTRICT"), nullable=False, index=True)
+    )
+    file_version_id: Mapped[str] = mapped_column(
+        ForeignKey("file_versions.id", ondelete="RESTRICT"), nullable=False, index=True
+    )
+    file_blob_id: Mapped[str] = mapped_column(
+        ForeignKey("file_blobs.id", ondelete="RESTRICT"), nullable=False, index=True
+    )
     filename_used: Mapped[str] = mapped_column(String(500), nullable=False)
     checksum_sha256: Mapped[str] = mapped_column(String(64), nullable=False)
     size_bytes: Mapped[int] = mapped_column(Integer, nullable=False)
     content_type: Mapped[str | None] = mapped_column(String(255))
-    use_stage: Mapped[str] = mapped_column(String(20), default="built", nullable=False, index=True)
+    use_stage: Mapped[str] = mapped_column(
-    used_at: Mapped[datetime | None] = mapped_column(DateTime(timezone=True), nullable=True, index=True)
+        String(20), default="built", nullable=False, index=True
+    )
+    used_at: Mapped[datetime | None] = mapped_column(
+        DateTime(timezone=True), nullable=True, index=True
+    )
 
 
 __all__ = [
@@ -249,6 +577,7 @@ __all__ = [
     "FileConnectorProfile",
     "FileConnectorSpace",
     "FileFolder",
+    "FileFormEvidenceGrant",
     "FileShare",
     "FileVersion",
 ]

src/govoplan_files/backend/documentation.py

@@ -12,7 +12,7 @@ from govoplan_core.core.modules import (
     DocumentationLink,
     DocumentationTopic,
 )
-from govoplan_files.backend.storage.archives import ZIP_UPLOAD_MAX_FILES
+from govoplan_files.backend.storage.archives import ARCHIVE_UPLOAD_MAX_ENTRIES
 from govoplan_files.backend.storage.access import user_group_ids
 from govoplan_files.backend.storage.connector_visibility import (
     connector_profile_usable_for_import,
@@ -22,6 +22,9 @@ from govoplan_files.backend.storage.connector_visibility import (
 
 _DEFAULT_UPLOAD_MAX_BYTES = 50 * 1024 * 1024
 _DEFAULT_ZIP_MAX_BYTES = 250 * 1024 * 1024
+_DEFAULT_ARCHIVE_MAX_EXPANDED_BYTES = 2 * 1024 * 1024 * 1024
+_DEFAULT_ARCHIVE_MAX_EXPANSION_RATIO = 100
+_DEFAULT_ARCHIVE_PREVIEW_TTL_SECONDS = 30 * 60
 _FILES_READ_SCOPE = "files:file:read"
 _FILES_UPLOAD_SCOPE = "files:file:upload"
 
@@ -42,11 +45,47 @@ def documentation_topics(
         "file_upload_zip_max_bytes",
         default=_DEFAULT_ZIP_MAX_BYTES,
     )
+    archive_expanded_limit = _configured_positive_int(
+        context.settings,
+        "file_archive_max_expanded_bytes",
+        default=_DEFAULT_ARCHIVE_MAX_EXPANDED_BYTES,
+    )
+    archive_entry_limit = _configured_positive_int(
+        context.settings,
+        "file_archive_max_entries",
+        default=ARCHIVE_UPLOAD_MAX_ENTRIES,
+    )
+    archive_ratio_limit = _configured_positive_int(
+        context.settings,
+        "file_archive_max_expansion_ratio",
+        default=_DEFAULT_ARCHIVE_MAX_EXPANSION_RATIO,
+    )
+    archive_preview_ttl = _configured_positive_int(
+        context.settings,
+        "file_archive_preview_ttl_seconds",
+        default=_DEFAULT_ARCHIVE_PREVIEW_TTL_SECONDS,
+    )
     topics: list[DocumentationTopic] = []
     if upload_limit is not None:
         topics.append(_upload_topic(upload_limit))
-    if upload_limit is not None and zip_limit is not None:
+    if (
-        topics.append(_zip_topic(upload_limit, zip_limit))
+        upload_limit is not None
+        and zip_limit is not None
+        and archive_expanded_limit is not None
+        and archive_entry_limit is not None
+        and archive_ratio_limit is not None
+        and archive_preview_ttl is not None
+    ):
+        topics.append(
+            _archive_topic(
+                upload_limit,
+                zip_limit,
+                archive_expanded_limit,
+                archive_entry_limit,
+                archive_ratio_limit,
+                archive_preview_ttl,
+            )
+        )
     topics.append(_connector_import_topic(context))
     return tuple(topics)
 
@@ -118,19 +157,29 @@ def _upload_topic(max_bytes: int) -> DocumentationTopic:
     )
 
 
-def _zip_topic(max_file_bytes: int, max_zip_bytes: int) -> DocumentationTopic:
+def _archive_topic(
+    max_file_bytes: int,
+    max_archive_request_bytes: int,
+    max_expanded_bytes: int,
+    max_entries: int,
+    max_expansion_ratio: int,
+    preview_ttl_seconds: int,
+) -> DocumentationTopic:
     member_limit = _format_byte_limit(max_file_bytes)
-    archive_limit = _format_byte_limit(max_zip_bytes)
+    request_limit = _format_byte_limit(max_archive_request_bytes)
+    expanded_limit = _format_byte_limit(max_expanded_bytes)
+    preview_minutes = max(1, preview_ttl_seconds // 60)
     return DocumentationTopic(
         id="files.workflow.upload-and-unpack-zip",
-        title="Upload and unpack a ZIP archive",
+        title="Preview and unpack an archive",
         summary=(
-            f"Safely unpack up to {ZIP_UPLOAD_MAX_FILES:,} files from a ZIP whose request and extracted total are each limited to {archive_limit}."
+            f"Review and selectively unpack up to {max_entries:,} entries from ZIP or TAR archives before any managed file is created."
         ),
         body=(
-            f"The current deployment limits the ZIP request and actual extracted total to {archive_limit}, each member to {member_limit}, "
+            f"The deployment accepts ZIP, TAR, TAR.GZ, TAR.BZ2, and TAR.XZ requests up to {request_limit}, limits actual expanded data to {expanded_limit}, "
-            f"and the archive to {ZIP_UPLOAD_MAX_FILES:,} non-directory members. Encrypted archives and unsafe member paths are rejected. "
+            f"each member to {member_limit}, the archive to {max_entries:,} entries, and expansion to {max_expansion_ratio}:1. "
-            "Actual extracted bytes are counted instead of trusting ZIP headers."
+            f"The server-issued preview expires after {preview_minutes} minutes. Password-protected ZIP archives are supported; passwords remain request-only. "
+            "Unsafe paths and special filesystem entries are rejected. Actual extracted bytes are counted instead of trusting archive headers."
         ),
         layer="configured",
         documentation_types=("user",),
@@ -161,35 +210,42 @@ def _zip_topic(max_file_bytes: int, max_zip_bytes: int) -> DocumentationTopic:
             "help_contexts": ["files.list"],
             "prerequisites": [
                 "You may view and upload managed files.",
-                "The archive is not encrypted and fits the current configured limits.",
+                "The archive fits the configured request, expansion, size, and entry limits.",
                 "The destination personal or group space grants this account write access.",
             ],
             "steps": [
                 "Open Files and choose the managed destination space and folder.",
-                "Enable Unpack ZIP uploads, then choose or drag the ZIP archive.",
+                "Enable Preview and unpack archive, then choose or drag one supported archive.",
+                "Review the discovered entries, supply a ZIP password when required, and select the files or folders to import.",
                 "Resolve every destination conflict explicitly.",
-                "Wait for extraction and finalization to finish before leaving the page.",
+                "Confirm the selection, then wait for extraction and finalization to finish before leaving the page.",
             ],
             "outcome": "Accepted archive members are stored as separate governed managed assets below the selected folder.",
             "verification": "Confirm the expected member paths and inspect representative file sizes, checksums, and versions.",
             "constraints": [
                 {
-                    "id": "zip-request-and-total",
+                    "id": "archive-request-and-total",
-                    "label": "Maximum ZIP request and extracted total",
+                    "label": "Maximum archive request and expanded total",
-                    "description": f"Both the compressed request and the actual extracted total are limited to {archive_limit}.",
+                    "description": f"The compressed request is limited to {request_limit}; actual expanded data is limited to {expanded_limit}.",
-                    "values": [archive_limit],
+                    "values": [request_limit, expanded_limit],
                 },
                 {
-                    "id": "zip-member-size",
+                    "id": "archive-member-size",
                     "label": "Maximum extracted member size",
                     "description": f"Each extracted file is limited to {member_limit}.",
                     "values": [member_limit],
                 },
                 {
-                    "id": "zip-member-count",
+                    "id": "archive-entry-count",
-                    "label": "Maximum file count",
+                    "label": "Maximum entry count",
-                    "description": f"A ZIP may contain at most {ZIP_UPLOAD_MAX_FILES:,} non-directory members.",
+                    "description": f"An archive may contain at most {max_entries:,} declared entries.",
-                    "values": [f"{ZIP_UPLOAD_MAX_FILES:,} files"],
+                    "values": [f"{max_entries:,} entries"],
+                },
+                {
+                    "id": "archive-expansion-ratio",
+                    "label": "Maximum expansion ratio",
+                    "description": f"Declared and actual output may not exceed {max_expansion_ratio} times the compressed request size.",
+                    "values": [f"{max_expansion_ratio}:1"],
                 },
             ],
             "related_topic_ids": [

src/govoplan_files/backend/dsar_provider.py

@@ -0,0 +1,879 @@
+from __future__ import annotations
+
+from collections.abc import Sequence
+from datetime import datetime, timezone
+
+from sqlalchemy import inspect, or_
+from sqlalchemy.orm import Session
+
+from govoplan_core.core.dsar import (
+    DsarErasureActionRef,
+    DsarExecutionResultRef,
+    DsarRecordRef,
+    DsarSubjectRef,
+    dsar_capability_name,
+)
+from govoplan_files.backend.db.models import (
+    CampaignAttachmentUse,
+    FileAsset,
+    FileConnectorCredential,
+    FileConnectorPolicy,
+    FileConnectorProfile,
+    FileConnectorSpace,
+    FileFolder,
+    FileFormEvidenceGrant,
+    FileIntegrityFinding,
+    FileIntegrityScan,
+    FileShare,
+    FileVersion,
+)
+
+
+FILES_DSAR_CAPABILITY = dsar_capability_name("files")
+_MAX_RECORDS = 5_000
+
+
+class FilesDsarProvider:
+    provider_id = "files"
+    module_id = "files"
+
+    def search_subject(
+        self,
+        session: object,
+        *,
+        tenant_id: str,
+        subject: DsarSubjectRef,
+    ) -> Sequence[DsarRecordRef]:
+        db = _session(session)
+        subject_user_id = _subject_user_id(subject)
+        if subject_user_id is None:
+            return ()
+
+        records: list[DsarRecordRef] = []
+        form_evidence_available = _has_table(db, FileFormEvidenceGrant)
+        campaign_evidence_available = _has_table(db, CampaignAttachmentUse)
+        retention_reasons: dict[str, str | None] = {}
+
+        def retention_reason(asset: FileAsset) -> str | None:
+            if asset.id not in retention_reasons:
+                retention_reasons[asset.id] = _asset_retention_reason(
+                    db,
+                    asset,
+                    form_evidence_available=form_evidence_available,
+                    campaign_evidence_available=campaign_evidence_available,
+                )
+            return retention_reasons[asset.id]
+
+        def append(record: DsarRecordRef) -> None:
+            if len(records) >= _MAX_RECORDS:
+                raise ValueError(
+                    "Files DSAR match limit exceeded; narrow the subject selectors."
+                )
+            records.append(record)
+
+        assets = _bounded_rows(
+            db.query(FileAsset)
+            .filter(
+                FileAsset.tenant_id == tenant_id,
+                or_(
+                    FileAsset.owner_user_id == subject_user_id,
+                    FileAsset.created_by_user_id == subject_user_id,
+                ),
+            )
+            .order_by(FileAsset.id)
+        )
+        for asset in assets:
+            match_fields = _matching_fields(
+                asset,
+                subject_user_id,
+                ("owner_user_id", "created_by_user_id"),
+            )
+            asset_retention_reason = retention_reason(asset)
+            append(
+                _record(
+                    "file_asset",
+                    asset.id,
+                    "managed_file",
+                    asset.filename,
+                    {
+                        "match_fields": match_fields,
+                        "owner_type": asset.owner_type,
+                        "display_path": asset.display_path,
+                        "filename": asset.filename,
+                        "description": asset.description,
+                        "deleted_at": _iso(asset.deleted_at),
+                        "retained_until": _iso(asset.retained_until),
+                        "legal_hold": asset.legal_hold,
+                        "lifecycle_reason": asset.lifecycle_reason,
+                        "lifecycle_revision": asset.lifecycle_revision,
+                    },
+                    observed_at=asset.updated_at,
+                    immutable=asset_retention_reason is not None,
+                    retention_reason=asset_retention_reason,
+                    source_path=f"/files?file={asset.id}",
+                )
+            )
+
+        version_rows = _bounded_rows(
+            db.query(FileVersion, FileAsset)
+            .join(FileAsset, FileAsset.id == FileVersion.file_asset_id)
+            .filter(
+                FileVersion.tenant_id == tenant_id,
+                FileAsset.tenant_id == tenant_id,
+                or_(
+                    FileAsset.owner_user_id == subject_user_id,
+                    FileVersion.created_by_user_id == subject_user_id,
+                ),
+            )
+            .order_by(FileVersion.id)
+        )
+        for version, asset in version_rows:
+            asset_retention_reason = retention_reason(asset)
+            append(
+                _record(
+                    "file_version",
+                    version.id,
+                    "managed_file_version",
+                    version.filename_at_upload,
+                    {
+                        "match_fields": (
+                            ["asset.owner_user_id"]
+                            if asset.owner_user_id == subject_user_id
+                            else []
+                        )
+                        + (
+                            ["created_by_user_id"]
+                            if version.created_by_user_id == subject_user_id
+                            else []
+                        ),
+                        "file_asset_id": version.file_asset_id,
+                        "version_number": version.version_number,
+                        "filename_at_upload": version.filename_at_upload,
+                        "display_path_at_upload": version.display_path_at_upload,
+                        "content_type": version.content_type,
+                        "size_bytes": version.size_bytes,
+                        "checksum_sha256": version.checksum_sha256,
+                        "created_at": _iso(version.created_at),
+                    },
+                    observed_at=version.updated_at,
+                    immutable=asset_retention_reason is not None,
+                    retention_reason=asset_retention_reason,
+                    source_path=(
+                        f"/api/v1/files/{version.file_asset_id}/versions/"
+                        f"{version.id}/download"
+                    ),
+                )
+            )
+
+        for folder in _bounded_rows(
+            db.query(FileFolder)
+            .filter(
+                FileFolder.tenant_id == tenant_id,
+                or_(
+                    FileFolder.owner_user_id == subject_user_id,
+                    FileFolder.created_by_user_id == subject_user_id,
+                ),
+            )
+            .order_by(FileFolder.id)
+        ):
+            append(
+                _record(
+                    "file_folder",
+                    folder.id,
+                    "managed_folder",
+                    folder.path,
+                    {
+                        "match_fields": _matching_fields(
+                            folder,
+                            subject_user_id,
+                            ("owner_user_id", "created_by_user_id"),
+                        ),
+                        "owner_type": folder.owner_type,
+                        "path": folder.path,
+                        "deleted_at": _iso(folder.deleted_at),
+                    },
+                    observed_at=folder.updated_at,
+                    source_path="/files",
+                )
+            )
+
+        for share in _bounded_rows(
+            db.query(FileShare)
+            .filter(
+                FileShare.tenant_id == tenant_id,
+                or_(
+                    (FileShare.target_type == "user")
+                    & (FileShare.target_id == subject_user_id),
+                    FileShare.created_by_user_id == subject_user_id,
+                    FileShare.revoked_by_user_id == subject_user_id,
+                ),
+            )
+            .order_by(FileShare.id)
+        ):
+            target_matches = (
+                share.target_type == "user" and share.target_id == subject_user_id
+            )
+            match_fields = _matching_fields(
+                share,
+                subject_user_id,
+                ("created_by_user_id", "revoked_by_user_id"),
+            )
+            if target_matches:
+                match_fields.insert(0, "target_id")
+            append(
+                _record(
+                    "file_share",
+                    share.id,
+                    "file_access_evidence",
+                    f"Share for file {share.file_asset_id}",
+                    {
+                        "match_fields": match_fields,
+                        "file_asset_id": share.file_asset_id,
+                        "target_type": share.target_type,
+                        "target_is_subject": target_matches,
+                        "permission": share.permission,
+                        "expires_at": _iso(share.expires_at),
+                        "revoked_at": _iso(share.revoked_at),
+                    },
+                    observed_at=share.updated_at,
+                    immutable=True,
+                    retention_reason=(
+                        "File-sharing history is institutional access evidence."
+                    ),
+                    source_path=f"/files?file={share.file_asset_id}",
+                )
+            )
+
+        if form_evidence_available:
+            evidence_rows = _bounded_rows(
+                db.query(FileFormEvidenceGrant)
+                .outerjoin(
+                    FileAsset,
+                    FileAsset.id == FileFormEvidenceGrant.file_asset_id,
+                )
+                .filter(
+                    FileFormEvidenceGrant.tenant_id == tenant_id,
+                    or_(
+                        FileFormEvidenceGrant.custodian_user_id == subject_user_id,
+                        FileAsset.owner_user_id == subject_user_id,
+                    ),
+                )
+                .order_by(FileFormEvidenceGrant.id)
+            )
+            for grant in evidence_rows:
+                append(
+                    _record(
+                        "file_form_evidence",
+                        grant.id,
+                        "form_evidence",
+                        f"Form evidence {grant.form_definition_id}",
+                        {
+                            "match_fields": (
+                                ["custodian_user_id"]
+                                if grant.custodian_user_id == subject_user_id
+                                else ["asset.owner_user_id"]
+                            ),
+                            "form_instance_id": grant.form_instance_id,
+                            "form_definition_id": grant.form_definition_id,
+                            "form_definition_revision": grant.form_definition_revision,
+                            "evidence_kind": grant.evidence_kind,
+                            "purpose": grant.purpose,
+                            "status": grant.status,
+                            "expires_at": _iso(grant.expires_at),
+                            "file_asset_id": grant.file_asset_id,
+                            "file_version_id": grant.file_version_id,
+                        },
+                        observed_at=grant.updated_at,
+                        immutable=True,
+                        retention_reason=(
+                            "Submitted Form attachment evidence follows the owning "
+                            "process retention and cannot be erased through Files alone."
+                        ),
+                    )
+                )
+
+        if campaign_evidence_available:
+            campaign_rows = _bounded_rows(
+                db.query(CampaignAttachmentUse)
+                .join(FileAsset, FileAsset.id == CampaignAttachmentUse.file_asset_id)
+                .filter(
+                    CampaignAttachmentUse.tenant_id == tenant_id,
+                    FileAsset.owner_user_id == subject_user_id,
+                )
+                .order_by(CampaignAttachmentUse.id)
+            )
+            for use in campaign_rows:
+                append(
+                    _record(
+                        "campaign_attachment_use",
+                        use.id,
+                        "delivery_evidence",
+                        use.filename_used,
+                        {
+                            "match_fields": ["asset.owner_user_id"],
+                            "campaign_id": use.campaign_id,
+                            "campaign_version_id": use.campaign_version_id,
+                            "campaign_job_id": use.campaign_job_id,
+                            "file_asset_id": use.file_asset_id,
+                            "file_version_id": use.file_version_id,
+                            "filename_used": use.filename_used,
+                            "checksum_sha256": use.checksum_sha256,
+                            "size_bytes": use.size_bytes,
+                            "use_stage": use.use_stage,
+                            "used_at": _iso(use.used_at),
+                        },
+                        observed_at=use.updated_at,
+                        immutable=True,
+                        retention_reason=(
+                            "Campaign attachment use is immutable delivery evidence."
+                        ),
+                    )
+                )
+
+        self._append_configuration_references(
+            db,
+            append=append,
+            tenant_id=tenant_id,
+            subject_user_id=subject_user_id,
+        )
+        self._append_integrity_references(
+            db,
+            append=append,
+            tenant_id=tenant_id,
+            subject_user_id=subject_user_id,
+        )
+        return tuple(records)
+
+    def plan_erasure(
+        self,
+        session: object,
+        *,
+        tenant_id: str,
+        subject: DsarSubjectRef,
+        records: Sequence[DsarRecordRef],
+    ) -> Sequence[DsarErasureActionRef]:
+        del session
+        subject_user_id = _subject_user_id(subject)
+        if subject_user_id is None:
+            return ()
+        actions: list[DsarErasureActionRef] = []
+        for record in records:
+            if record.provider_id != self.provider_id or record.module_id != self.module_id:
+                raise ValueError("Files DSAR received a foreign provider record.")
+            match_fields = {
+                str(value) for value in record.data.get("match_fields", ())
+            }
+            if record.immutable_evidence:
+                actions.append(
+                    _action(
+                        f"files:retain:{record.resource_type}:{record.resource_id}",
+                        "retain",
+                        record,
+                        f"Retain {record.title}",
+                        record.retention_reason
+                        or "Institutional evidence must be retained.",
+                        executable=False,
+                    )
+                )
+            elif record.resource_type in {
+                "file_asset",
+                "file_version",
+                "file_folder",
+            }:
+                actions.append(
+                    _action(
+                        f"files:review:{record.resource_type}:{record.resource_id}",
+                        "manual_review",
+                        record,
+                        f"Review {record.title}",
+                        (
+                            "Managed file content, names, paths, ownership, and shared "
+                            "references require a case decision. Approved byte erasure "
+                            "must use the separate Files purge workflow."
+                        ),
+                        executable=False,
+                    )
+                )
+            if (
+                record.resource_type == "file_share"
+                and "target_id" in match_fields
+                and record.data.get("revoked_at") is None
+            ):
+                actions.append(
+                    _action(
+                        f"files:revoke:file_share:{record.resource_id}",
+                        "revoke",
+                        record,
+                        "Revoke active file share",
+                        "The active user-targeted share can be revoked without deleting file evidence.",
+                        executable=True,
+                        metadata={"subject_user_id": subject_user_id},
+                    )
+                )
+            for field_name in sorted(
+                match_fields.intersection(_detachable_fields(record.resource_type))
+            ):
+                actions.append(
+                    _action(
+                        (
+                            f"files:detach:{record.resource_type}:"
+                            f"{field_name}:{record.resource_id}"
+                        ),
+                        "detach",
+                        record,
+                        f"Detach {field_name.replace('_', ' ')}",
+                        (
+                            "Remove the mutable subject reference while preserving the "
+                            "governed resource and DSAR evidence."
+                        ),
+                        executable=True,
+                        metadata={
+                            "field": field_name,
+                            "subject_user_id": subject_user_id,
+                        },
+                    )
+                )
+        action_ids = [action.action_id for action in actions]
+        if len(action_ids) != len(set(action_ids)):
+            raise ValueError("Files DSAR produced duplicate action ids.")
+        return tuple(actions)
+
+    def execute_erasure(
+        self,
+        session: object,
+        *,
+        tenant_id: str,
+        subject: DsarSubjectRef,
+        actions: Sequence[DsarErasureActionRef],
+        request_id: str,
+    ) -> Sequence[DsarExecutionResultRef]:
+        db = _session(session)
+        subject_user_id = _subject_user_id(subject)
+        if subject_user_id is None:
+            return tuple(
+                _blocked(action, "Files requires a direct membership subject reference.")
+                for action in actions
+            )
+        results: list[DsarExecutionResultRef] = []
+        for action in actions:
+            if (
+                action.provider_id != self.provider_id
+                or action.module_id != self.module_id
+                or action.metadata.get("subject_user_id") != subject_user_id
+            ):
+                results.append(_blocked(action, "The Files DSAR action is stale or invalid."))
+                continue
+            if action.action_id.startswith("files:revoke:file_share:"):
+                results.append(
+                    _revoke_share(
+                        db,
+                        tenant_id=tenant_id,
+                        subject_user_id=subject_user_id,
+                        action=action,
+                        request_id=request_id,
+                    )
+                )
+            elif action.action_id.startswith("files:detach:"):
+                results.append(
+                    _detach_reference(
+                        db,
+                        tenant_id=tenant_id,
+                        subject_user_id=subject_user_id,
+                        action=action,
+                        request_id=request_id,
+                    )
+                )
+            else:
+                results.append(_blocked(action, "Files does not execute this action kind."))
+        db.flush()
+        return tuple(results)
+
+    def _append_configuration_references(
+        self,
+        db: Session,
+        *,
+        append: object,
+        tenant_id: str,
+        subject_user_id: str,
+    ) -> None:
+        configurations = (
+            (FileConnectorProfile, "connector_profile", ("created_by_user_id", "updated_by_user_id")),
+            (FileConnectorCredential, "connector_credential", ("created_by_user_id", "updated_by_user_id")),
+            (FileConnectorPolicy, "connector_policy", ("created_by_user_id", "updated_by_user_id")),
+            (FileConnectorSpace, "connector_space", ("owner_user_id", "created_by_user_id")),
+        )
+        for model, resource_type, fields in configurations:
+            if not _has_table(db, model):
+                continue
+            conditions = [getattr(model, field) == subject_user_id for field in fields]
+            query = db.query(model).filter(or_(*conditions))
+            if hasattr(model, "tenant_id"):
+                query = query.filter(model.tenant_id == tenant_id)
+            for row in _bounded_rows(query.order_by(model.id)):
+                match_fields = _matching_fields(row, subject_user_id, fields)
+                data: dict[str, object] = {
+                    "match_fields": match_fields,
+                    "label": getattr(row, "label", None),
+                    "provider": getattr(row, "provider", None),
+                }
+                if isinstance(row, FileConnectorCredential):
+                    data["credential_mode"] = row.credential_mode
+                elif isinstance(row, FileConnectorSpace):
+                    data.update(
+                        {
+                            "remote_path": row.remote_path,
+                            "sync_mode": row.sync_mode,
+                            "read_only": row.read_only,
+                            "deleted_at": _iso(row.deleted_at),
+                        }
+                    )
+                append(  # type: ignore[operator]
+                    _record(
+                        resource_type,
+                        row.id,
+                        "connector_configuration_evidence",
+                        getattr(row, "label", None) or resource_type.replace("_", " "),
+                        data,
+                        observed_at=row.updated_at,
+                        immutable=True,
+                        retention_reason=(
+                            "Connector configuration history is institutional evidence; "
+                            "credential secrets are excluded from the DSAR export."
+                        ),
+                    )
+                )
+
+    def _append_integrity_references(
+        self,
+        db: Session,
+        *,
+        append: object,
+        tenant_id: str,
+        subject_user_id: str,
+    ) -> None:
+        if _has_table(db, FileIntegrityScan):
+            for scan in _bounded_rows(
+                db.query(FileIntegrityScan)
+                .filter(
+                    FileIntegrityScan.tenant_id == tenant_id,
+                    FileIntegrityScan.created_by_user_id == subject_user_id,
+                )
+                .order_by(FileIntegrityScan.id)
+            ):
+                append(  # type: ignore[operator]
+                    _record(
+                        "file_integrity_scan",
+                        scan.id,
+                        "storage_integrity_evidence",
+                        f"Integrity scan {scan.id}",
+                        {
+                            "match_fields": ["created_by_user_id"],
+                            "storage_backend": scan.storage_backend,
+                            "status": scan.status,
+                            "started_at": _iso(scan.started_at),
+                            "completed_at": _iso(scan.completed_at),
+                        },
+                        observed_at=scan.updated_at,
+                        immutable=True,
+                        retention_reason="Storage integrity scans are operator evidence.",
+                    )
+                )
+        if _has_table(db, FileIntegrityFinding):
+            for finding in _bounded_rows(
+                db.query(FileIntegrityFinding)
+                .filter(
+                    FileIntegrityFinding.tenant_id == tenant_id,
+                    FileIntegrityFinding.resolved_by_user_id == subject_user_id,
+                )
+                .order_by(FileIntegrityFinding.id)
+            ):
+                append(  # type: ignore[operator]
+                    _record(
+                        "file_integrity_finding",
+                        finding.id,
+                        "storage_integrity_evidence",
+                        f"Integrity finding {finding.kind}",
+                        {
+                            "match_fields": ["resolved_by_user_id"],
+                            "kind": finding.kind,
+                            "state": finding.state,
+                            "resolved_at": _iso(finding.resolved_at),
+                        },
+                        observed_at=finding.updated_at,
+                        immutable=True,
+                        retention_reason="Integrity resolution is operator evidence.",
+                    )
+                )
+
+
+def _asset_retention_reason(
+    session: Session,
+    asset: FileAsset,
+    *,
+    form_evidence_available: bool,
+    campaign_evidence_available: bool,
+) -> str | None:
+    reasons: list[str] = []
+    if asset.legal_hold:
+        reasons.append("The file is under legal hold.")
+    retained_until = _aware(asset.retained_until)
+    if retained_until is not None and retained_until > datetime.now(timezone.utc):
+        reasons.append(f"The file is retained until {retained_until.isoformat()}.")
+    if form_evidence_available:
+        has_form_evidence = (
+            session.query(FileFormEvidenceGrant.id)
+            .filter(
+                FileFormEvidenceGrant.tenant_id == asset.tenant_id,
+                FileFormEvidenceGrant.file_asset_id == asset.id,
+            )
+            .first()
+            is not None
+        )
+        if has_form_evidence:
+            reasons.append("The file is referenced by submitted Form evidence.")
+    if campaign_evidence_available:
+        has_campaign_evidence = (
+            session.query(CampaignAttachmentUse.id)
+            .filter(
+                CampaignAttachmentUse.tenant_id == asset.tenant_id,
+                CampaignAttachmentUse.file_asset_id == asset.id,
+            )
+            .first()
+            is not None
+        )
+        if has_campaign_evidence:
+            reasons.append("The file is referenced by Campaign delivery evidence.")
+    return " ".join(reasons) or None
+
+
+_DETACHABLE_MODELS: dict[str, tuple[type[object], frozenset[str]]] = {
+    "file_asset": (FileAsset, frozenset({"created_by_user_id"})),
+    "file_version": (FileVersion, frozenset({"created_by_user_id"})),
+    "file_folder": (FileFolder, frozenset({"created_by_user_id"})),
+    "file_share": (
+        FileShare,
+        frozenset({"created_by_user_id", "revoked_by_user_id"}),
+    ),
+    "connector_profile": (
+        FileConnectorProfile,
+        frozenset({"created_by_user_id", "updated_by_user_id"}),
+    ),
+    "connector_credential": (
+        FileConnectorCredential,
+        frozenset({"created_by_user_id", "updated_by_user_id"}),
+    ),
+    "connector_policy": (
+        FileConnectorPolicy,
+        frozenset({"created_by_user_id", "updated_by_user_id"}),
+    ),
+    "connector_space": (FileConnectorSpace, frozenset({"created_by_user_id"})),
+}
+
+
+def _detachable_fields(resource_type: str) -> frozenset[str]:
+    entry = _DETACHABLE_MODELS.get(resource_type)
+    return entry[1] if entry is not None else frozenset()
+
+
+def _revoke_share(
+    session: Session,
+    *,
+    tenant_id: str,
+    subject_user_id: str,
+    action: DsarErasureActionRef,
+    request_id: str,
+) -> DsarExecutionResultRef:
+    row = (
+        session.query(FileShare)
+        .filter(FileShare.id == action.resource_id)
+        .with_for_update()
+        .one_or_none()
+    )
+    if (
+        row is None
+        or row.tenant_id != tenant_id
+        or row.target_type != "user"
+        or row.target_id != subject_user_id
+    ):
+        return _blocked(action, "The subject-targeted file share is no longer available.")
+    if row.revoked_at is not None:
+        return _result(
+            action,
+            "unchanged",
+            "The file share was already revoked.",
+            {"request_id": request_id, "revoked_at": _iso(row.revoked_at)},
+        )
+    row.revoked_at = datetime.now(timezone.utc)
+    row.revoked_by_user_id = None
+    return _result(
+        action,
+        "executed",
+        "The subject-targeted file share was revoked.",
+        {"request_id": request_id, "revoked_at": _iso(row.revoked_at)},
+    )
+
+
+def _detach_reference(
+    session: Session,
+    *,
+    tenant_id: str,
+    subject_user_id: str,
+    action: DsarErasureActionRef,
+    request_id: str,
+) -> DsarExecutionResultRef:
+    entry = _DETACHABLE_MODELS.get(action.resource_type)
+    field_name = str(action.metadata.get("field") or "")
+    if entry is None or field_name not in entry[1]:
+        return _blocked(action, "The requested Files subject reference is not detachable.")
+    model = entry[0]
+    row = (
+        session.query(model)
+        .filter(getattr(model, "id") == action.resource_id)
+        .with_for_update()
+        .one_or_none()
+    )
+    if row is None or getattr(row, "tenant_id", None) != tenant_id:
+        return _blocked(action, "The Files resource is no longer available.")
+    current = getattr(row, field_name)
+    if current is None:
+        return _result(
+            action,
+            "unchanged",
+            "The subject reference was already detached.",
+            {"request_id": request_id, "field": field_name},
+        )
+    if current != subject_user_id:
+        return _blocked(action, "The Files subject reference changed after planning.")
+    setattr(row, field_name, None)
+    return _result(
+        action,
+        "executed",
+        "The mutable subject reference was detached.",
+        {"request_id": request_id, "field": field_name},
+    )
+
+
+def _subject_user_id(subject: DsarSubjectRef) -> str | None:
+    candidates: list[str] = []
+    if subject.membership_id:
+        candidates.append(subject.membership_id)
+    for key, value in subject.external_references.items():
+        if key in {
+            "files.user",
+            "files.membership",
+            "access.membership",
+            "membership_id",
+        }:
+            candidates.append(value)
+    normalized = {value.strip() for value in candidates if value.strip()}
+    if len(normalized) != 1:
+        return None
+    return normalized.pop()
+
+
+def _matching_fields(
+    row: object,
+    subject_user_id: str,
+    fields: Sequence[str],
+) -> list[str]:
+    return [field for field in fields if getattr(row, field) == subject_user_id]
+
+
+def _record(
+    resource_type: str,
+    resource_id: str,
+    category: str,
+    title: str,
+    data: dict[str, object],
+    *,
+    observed_at: datetime | None = None,
+    immutable: bool = False,
+    retention_reason: str | None = None,
+    source_path: str | None = None,
+) -> DsarRecordRef:
+    return DsarRecordRef(
+        provider_id="files",
+        module_id="files",
+        resource_type=resource_type,
+        resource_id=resource_id,
+        category=category,
+        title=title,
+        data=data,
+        observed_at=observed_at,
+        immutable_evidence=immutable,
+        retention_reason=retention_reason,
+        source_path=source_path,
+    )
+
+
+def _action(
+    action_id: str,
+    kind: str,
+    record: DsarRecordRef,
+    title: str,
+    rationale: str,
+    *,
+    executable: bool,
+    metadata: dict[str, object] | None = None,
+) -> DsarErasureActionRef:
+    return DsarErasureActionRef(
+        action_id=action_id,
+        provider_id="files",
+        module_id="files",
+        kind=kind,  # type: ignore[arg-type]
+        resource_type=record.resource_type,
+        resource_id=record.resource_id,
+        title=title,
+        rationale=rationale,
+        executable=executable,
+        metadata=metadata or {},
+    )
+
+
+def _result(
+    action: DsarErasureActionRef,
+    status: str,
+    summary: str,
+    evidence: dict[str, object] | None = None,
+) -> DsarExecutionResultRef:
+    return DsarExecutionResultRef(
+        action_id=action.action_id,
+        status=status,  # type: ignore[arg-type]
+        summary=summary,
+        evidence=evidence or {},
+    )
+
+
+def _blocked(action: DsarErasureActionRef, summary: str) -> DsarExecutionResultRef:
+    return _result(action, "blocked", summary)
+
+
+def _session(value: object) -> Session:
+    if not isinstance(value, Session):
+        raise TypeError("Files DSAR provider requires a SQLAlchemy session.")
+    return value
+
+
+def _bounded_rows(query: object) -> list[object]:
+    rows = query.limit(_MAX_RECORDS + 1).all()  # type: ignore[attr-defined]
+    if len(rows) > _MAX_RECORDS:
+        raise ValueError("Files DSAR match limit exceeded; narrow the subject selectors.")
+    return rows
+
+
+def _has_table(session: Session, model: type[object]) -> bool:
+    return inspect(session.connection()).has_table(model.__tablename__)
+
+
+def _aware(value: datetime | None) -> datetime | None:
+    if value is None or value.tzinfo is not None:
+        return value
+    return value.replace(tzinfo=timezone.utc)
+
+
+def _iso(value: datetime | None) -> str | None:
+    aware = _aware(value)
+    return aware.isoformat() if aware else None
+
+
+__all__ = ["FILES_DSAR_CAPABILITY", "FilesDsarProvider"]

src/govoplan_files/backend/form_evidence.py

@@ -0,0 +1,449 @@
+from __future__ import annotations
+
+from collections.abc import Mapping, Sequence
+from datetime import UTC, datetime, timedelta
+import hashlib
+import json
+import secrets
+
+from sqlalchemy.orm import Session
+
+from govoplan_core.core.access import CAPABILITY_ACCESS_DIRECTORY, AccessDirectory
+from govoplan_core.core.form_evidence import (
+    FormEvidenceContractError,
+    FormEvidenceGrant,
+    FormEvidenceGrantRequest,
+    FormEvidenceInspection,
+    FormEvidenceInspectionRequest,
+    FormEvidenceState,
+)
+from govoplan_core.core.institutional import EvidenceReference
+from govoplan_core.core.modules import ModuleContext
+from govoplan_files.backend.db.models import (
+    FileAsset,
+    FileBlob,
+    FileFormEvidenceGrant,
+    FileVersion,
+)
+from govoplan_files.backend.runtime import configure_runtime
+
+
+PROVIDER_ID = "files"
+CAPABILITY_FORM_EVIDENCE_FILES = "forms_runtime.evidence.files"
+MAX_GRANT_TTL = timedelta(minutes=15)
+
+
+class FilesFormEvidenceProvider:
+    provider_id = PROVIDER_ID
+
+    def __init__(self, registry: object | None, settings: object) -> None:
+        self._registry = registry
+        self._settings = settings
+
+    def supported_kinds(self) -> Sequence[str]:
+        return ("document",)
+
+    def create_upload_grant(
+        self,
+        session: object,
+        principal: object,
+        *,
+        request: FormEvidenceGrantRequest,
+    ) -> FormEvidenceGrant:
+        db = _session(session)
+        _assert_tenant(principal, request.tenant_id)
+        if request.evidence_kind != "document":
+            raise FormEvidenceContractError(
+                "Files can accept only document evidence for Forms Runtime."
+            )
+        custodian_user_id = _custodian_user_id(request.custodian_ref)
+        self._assert_active_custodian(
+            tenant_id=request.tenant_id,
+            user_id=custodian_user_id,
+        )
+        now = datetime.now(UTC)
+        if request.expires_at <= now:
+            raise FormEvidenceContractError(
+                "Form evidence upload grant expiry must be in the future."
+            )
+        allowed_content_types = _content_types(request.allowed_content_types)
+        configured_max = int(getattr(self._settings, "file_upload_max_bytes"))
+        max_size_bytes = min(request.max_size_bytes or configured_max, configured_max)
+        request_sha256 = _request_sha256(
+            request,
+            custodian_user_id=custodian_user_id,
+            max_size_bytes=max_size_bytes,
+            allowed_content_types=allowed_content_types,
+        )
+        existing = (
+            db.query(FileFormEvidenceGrant)
+            .filter(
+                FileFormEvidenceGrant.tenant_id == request.tenant_id,
+                FileFormEvidenceGrant.idempotency_key == request.idempotency_key,
+            )
+            .one_or_none()
+        )
+        if existing is not None:
+            if not secrets.compare_digest(existing.request_sha256, request_sha256):
+                raise FormEvidenceContractError(
+                    "Form evidence grant idempotency conflict."
+                )
+            if (
+                existing.status in {"expired", "revoked"}
+                or _aware(existing.expires_at) <= now
+            ):
+                raise FormEvidenceContractError(
+                    "The existing Form evidence upload grant is no longer usable; "
+                    "request a new grant with a new idempotency key."
+                )
+            return _grant_response(existing, upload_token=None, replayed=True)
+
+        metadata = _bounded_metadata(request.metadata)
+        remaining_attachments = metadata.get("remaining_attachments")
+        if isinstance(remaining_attachments, int):
+            existing_attachment_ids = set(metadata.get("existing_attachment_ids", ()))
+            active_grants = (
+                db.query(FileFormEvidenceGrant)
+                .filter(
+                    FileFormEvidenceGrant.tenant_id == request.tenant_id,
+                    FileFormEvidenceGrant.form_instance_id == request.instance_id,
+                    FileFormEvidenceGrant.form_definition_id
+                    == request.definition_ref.object_id,
+                    FileFormEvidenceGrant.form_definition_revision
+                    == str(request.definition_ref.version),
+                    FileFormEvidenceGrant.status.in_(("issued", "uploaded")),
+                    FileFormEvidenceGrant.expires_at > now,
+                )
+                .with_for_update()
+                .all()
+            )
+            outstanding = sum(
+                1
+                for item in active_grants
+                if not item.file_asset_id
+                or item.file_asset_id not in existing_attachment_ids
+            )
+            if outstanding >= remaining_attachments:
+                raise FormEvidenceContractError(
+                    "This Form already has the maximum number of active attachment uploads."
+                )
+
+        upload_token = secrets.token_urlsafe(32)
+        grant = FileFormEvidenceGrant(
+            tenant_id=request.tenant_id,
+            form_instance_id=request.instance_id,
+            form_definition_id=request.definition_ref.object_id,
+            form_definition_revision=str(request.definition_ref.version),
+            token_sha256=_token_sha256(upload_token),
+            idempotency_key=request.idempotency_key,
+            request_sha256=request_sha256,
+            custodian_user_id=custodian_user_id,
+            evidence_kind=request.evidence_kind,
+            purpose=request.purpose,
+            status="issued",
+            expires_at=min(request.expires_at, now + MAX_GRANT_TTL),
+            max_size_bytes=max_size_bytes,
+            allowed_content_types=list(allowed_content_types),
+            metadata_=metadata,
+        )
+        db.add(grant)
+        db.flush()
+        return _grant_response(grant, upload_token=upload_token, replayed=False)
+
+    def inspect_evidence(
+        self,
+        session: object,
+        principal: object,
+        *,
+        request: FormEvidenceInspectionRequest,
+    ) -> FormEvidenceInspection:
+        db = _session(session)
+        _assert_tenant(principal, request.tenant_id)
+        now = datetime.now(UTC)
+        if request.evidence.owner_module != PROVIDER_ID:
+            return _inspection(
+                request.evidence,
+                state="rejected",
+                observed_at=now,
+                reason="The evidence owner does not match the Files provider.",
+            )
+        if request.evidence.kind != "document" or not request.evidence.version:
+            return _inspection(
+                request.evidence,
+                state="rejected",
+                observed_at=now,
+                reason="Files evidence requires an exact document version.",
+            )
+        grant = (
+            db.query(FileFormEvidenceGrant)
+            .filter(
+                FileFormEvidenceGrant.tenant_id == request.tenant_id,
+                FileFormEvidenceGrant.form_instance_id == request.instance_id,
+                FileFormEvidenceGrant.form_definition_id
+                == request.definition_ref.object_id,
+                FileFormEvidenceGrant.form_definition_revision
+                == str(request.definition_ref.version),
+                FileFormEvidenceGrant.file_asset_id == request.evidence.evidence_id,
+                FileFormEvidenceGrant.file_version_id == request.evidence.version,
+            )
+            .one_or_none()
+        )
+        if grant is None:
+            return _inspection(
+                request.evidence,
+                state="rejected",
+                observed_at=now,
+                reason="The document was not captured for this exact Form submission.",
+            )
+        if grant.status == "revoked":
+            return _inspection(
+                request.evidence,
+                state="revoked",
+                observed_at=now,
+                reason="The Form evidence grant was revoked.",
+            )
+        if grant.status != "uploaded":
+            return _inspection(
+                request.evidence,
+                state="pending",
+                observed_at=now,
+                retryable=True,
+                reason="The Form evidence upload has not completed.",
+            )
+        asset = db.get(FileAsset, grant.file_asset_id)
+        version = db.get(FileVersion, grant.file_version_id)
+        blob = db.get(FileBlob, version.blob_id) if version is not None else None
+        if asset is None or version is None or blob is None:
+            return _inspection(
+                request.evidence,
+                state="unavailable",
+                observed_at=now,
+                retryable=True,
+                reason="The managed document cannot currently be reconstructed.",
+            )
+        if (
+            asset.tenant_id != request.tenant_id
+            or version.tenant_id != request.tenant_id
+            or blob.tenant_id != request.tenant_id
+            or version.file_asset_id != asset.id
+        ):
+            return _inspection(
+                request.evidence,
+                state="rejected",
+                observed_at=now,
+                reason="The managed document crosses an evidence ownership boundary.",
+            )
+        if asset.deleted_at is not None:
+            return _inspection(
+                request.evidence,
+                state="revoked",
+                observed_at=now,
+                reason="The managed document is no longer active.",
+            )
+        if blob.quarantined_at is not None or blob.integrity_status == "quarantined":
+            return _inspection(
+                request.evidence,
+                state="rejected",
+                observed_at=now,
+                reason="The managed document failed its integrity gate.",
+            )
+        if blob.integrity_status != "verified":
+            return _inspection(
+                request.evidence,
+                state="pending",
+                observed_at=now,
+                retryable=True,
+                reason="The managed document is awaiting integrity verification.",
+            )
+        if (
+            not request.evidence.checksum
+            or not secrets.compare_digest(
+                request.evidence.checksum,
+                version.checksum_sha256,
+            )
+            or not secrets.compare_digest(
+                version.checksum_sha256,
+                blob.checksum_sha256,
+            )
+        ):
+            return _inspection(
+                request.evidence,
+                state="rejected",
+                observed_at=now,
+                reason="The managed document checksum does not match the evidence.",
+            )
+        return _inspection(
+            request.evidence,
+            state="accepted",
+            observed_at=now,
+            metadata={
+                "content_type": version.content_type,
+                "size_bytes": version.size_bytes,
+                "integrity_status": blob.integrity_status,
+                "grant_id": grant.id,
+            },
+        )
+
+    def _assert_active_custodian(self, *, tenant_id: str, user_id: str) -> None:
+        registry = self._registry
+        if registry is None or not hasattr(registry, "has_capability"):
+            raise FormEvidenceContractError(
+                "The Access directory is unavailable for Form evidence custody."
+            )
+        if not registry.has_capability(CAPABILITY_ACCESS_DIRECTORY):
+            raise FormEvidenceContractError(
+                "The Access directory is unavailable for Form evidence custody."
+            )
+        directory = registry.require_capability(CAPABILITY_ACCESS_DIRECTORY)
+        if not isinstance(directory, AccessDirectory):
+            raise FormEvidenceContractError(
+                "The Access directory capability is invalid."
+            )
+        user = directory.get_user(user_id)
+        if user is None or user.tenant_id != tenant_id or user.status != "active":
+            raise FormEvidenceContractError(
+                "Form evidence requires an active same-tenant custodian."
+            )
+
+
+def create_files_form_evidence_provider(
+    context: ModuleContext,
+) -> FilesFormEvidenceProvider:
+    configure_runtime(registry=context.registry, settings=context.settings)
+    return FilesFormEvidenceProvider(context.registry, context.settings)
+
+
+def _grant_response(
+    grant: FileFormEvidenceGrant,
+    *,
+    upload_token: str | None,
+    replayed: bool,
+) -> FormEvidenceGrant:
+    return FormEvidenceGrant(
+        provider_id=PROVIDER_ID,
+        grant_id=grant.id,
+        upload_token=upload_token,
+        upload_url="/api/v1/files/form-evidence/upload",
+        expires_at=_aware(grant.expires_at),
+        max_size_bytes=grant.max_size_bytes,
+        allowed_content_types=tuple(grant.allowed_content_types),
+        replayed=replayed,
+    )
+
+
+def _inspection(
+    reference: EvidenceReference,
+    *,
+    state: FormEvidenceState,
+    observed_at: datetime,
+    retryable: bool = False,
+    reason: str | None = None,
+    metadata: Mapping[str, object] | None = None,
+) -> FormEvidenceInspection:
+    return FormEvidenceInspection(
+        provider_id=PROVIDER_ID,
+        reference=reference,
+        state=state,
+        observed_at=observed_at,
+        retryable=retryable,
+        reason=reason,
+        metadata=dict(metadata or {}),
+    )
+
+
+def _session(value: object) -> Session:
+    if not isinstance(value, Session):
+        raise TypeError("Files Form evidence requires a SQLAlchemy session.")
+    return value
+
+
+def _assert_tenant(principal: object, tenant_id: str) -> None:
+    principal_tenant = str(getattr(principal, "tenant_id", "") or "").strip()
+    if not principal_tenant or principal_tenant != tenant_id:
+        raise PermissionError("Form evidence cannot cross tenants.")
+
+
+def _custodian_user_id(value: str | None) -> str:
+    clean = str(value or "").strip()
+    if not clean.startswith("user:") or len(clean) <= len("user:"):
+        raise FormEvidenceContractError(
+            "Files Form evidence requires a user custodian."
+        )
+    return clean.removeprefix("user:")
+
+
+def _content_types(values: Sequence[str]) -> tuple[str, ...]:
+    cleaned = tuple(
+        dict.fromkeys(
+            value.split(";", 1)[0].strip().casefold()
+            for value in values
+            if value.strip()
+        )
+    )
+    if len(cleaned) > 50 or any(len(value) > 255 for value in cleaned):
+        raise FormEvidenceContractError(
+            "Form evidence content-type restrictions are too large."
+        )
+    return cleaned
+
+
+def _bounded_metadata(value: Mapping[str, object]) -> dict[str, object]:
+    remaining = value.get("remaining_attachments")
+    result: dict[str, object] = {}
+    if isinstance(remaining, int) and remaining >= 0:
+        result["remaining_attachments"] = remaining
+    raw_ids = value.get("existing_attachment_ids")
+    if isinstance(raw_ids, (list, tuple)):
+        clean_ids = tuple(
+            dict.fromkeys(
+                item.strip()
+                for item in raw_ids
+                if isinstance(item, str) and item.strip() and len(item.strip()) <= 255
+            )
+        )
+        if len(clean_ids) <= 1000:
+            result["existing_attachment_ids"] = clean_ids
+    return result
+
+
+def _request_sha256(
+    request: FormEvidenceGrantRequest,
+    *,
+    custodian_user_id: str,
+    max_size_bytes: int,
+    allowed_content_types: Sequence[str],
+) -> str:
+    return hashlib.sha256(
+        json.dumps(
+            {
+                "tenant_id": request.tenant_id,
+                "instance_id": request.instance_id,
+                "definition_ref": request.definition_ref.to_dict(),
+                "evidence_kind": request.evidence_kind,
+                "purpose": request.purpose,
+                "expires_at": request.expires_at.isoformat(),
+                "custodian_user_id": custodian_user_id,
+                "max_size_bytes": max_size_bytes,
+                "allowed_content_types": list(allowed_content_types),
+                "metadata": _bounded_metadata(request.metadata),
+            },
+            sort_keys=True,
+            separators=(",", ":"),
+        ).encode("utf-8")
+    ).hexdigest()
+
+
+def _token_sha256(token: str) -> str:
+    return hashlib.sha256(token.encode("utf-8")).hexdigest()
+
+
+def _aware(value: datetime) -> datetime:
+    return value if value.tzinfo is not None else value.replace(tzinfo=UTC)
+
+
+__all__ = [
+    "CAPABILITY_FORM_EVIDENCE_FILES",
+    "FilesFormEvidenceProvider",
+    "PROVIDER_ID",
+    "create_files_form_evidence_provider",
+]

src/govoplan_files/backend/migrations/versions/a2b3c4d5e6f8_form_evidence_upload_grants.py

@@ -0,0 +1,147 @@
+"""add purpose-bound Form evidence upload grants
+
+Revision ID: a2b3c4d5e6f8
+Revises: f1a2b3c4d5e7
+"""
+
+from __future__ import annotations
+
+import sqlalchemy as sa
+from alembic import op
+
+
+revision = "a2b3c4d5e6f8"
+down_revision = "f1a2b3c4d5e7"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    op.create_table(
+        "file_form_evidence_grants",
+        sa.Column("id", sa.String(length=36), nullable=False),
+        sa.Column("tenant_id", sa.String(length=36), nullable=False),
+        sa.Column("form_instance_id", sa.String(length=36), nullable=False),
+        sa.Column("form_definition_id", sa.String(length=255), nullable=False),
+        sa.Column("form_definition_revision", sa.String(length=255), nullable=False),
+        sa.Column("token_sha256", sa.String(length=64), nullable=False),
+        sa.Column("idempotency_key", sa.String(length=255), nullable=False),
+        sa.Column("request_sha256", sa.String(length=64), nullable=False),
+        sa.Column("custodian_user_id", sa.String(length=36), nullable=False),
+        sa.Column("evidence_kind", sa.String(length=30), nullable=False),
+        sa.Column("purpose", sa.String(length=500), nullable=False),
+        sa.Column("status", sa.String(length=30), nullable=False),
+        sa.Column("expires_at", sa.DateTime(timezone=True), nullable=False),
+        sa.Column("max_size_bytes", sa.Integer(), nullable=False),
+        sa.Column("allowed_content_types", sa.JSON(), nullable=False),
+        sa.Column("file_asset_id", sa.String(length=36), nullable=True),
+        sa.Column("file_version_id", sa.String(length=36), nullable=True),
+        sa.Column("uploaded_at", sa.DateTime(timezone=True), nullable=True),
+        sa.Column("revoked_at", sa.DateTime(timezone=True), nullable=True),
+        sa.Column("metadata", sa.JSON(), nullable=False),
+        sa.Column("created_at", sa.DateTime(timezone=True), nullable=False),
+        sa.Column("updated_at", sa.DateTime(timezone=True), nullable=False),
+        sa.ForeignKeyConstraint(
+            ["custodian_user_id"],
+            ["access_users.id"],
+            ondelete="RESTRICT",
+        ),
+        sa.ForeignKeyConstraint(
+            ["file_asset_id"],
+            ["file_assets.id"],
+            ondelete="RESTRICT",
+        ),
+        sa.ForeignKeyConstraint(
+            ["file_version_id"],
+            ["file_versions.id"],
+            ondelete="RESTRICT",
+        ),
+        sa.PrimaryKeyConstraint("id"),
+        sa.UniqueConstraint(
+            "tenant_id",
+            "idempotency_key",
+            name="uq_file_form_evidence_grants_idempotency",
+        ),
+        sa.UniqueConstraint("token_sha256"),
+    )
+    op.create_index(
+        "ix_file_form_evidence_grants_tenant_id",
+        "file_form_evidence_grants",
+        ["tenant_id"],
+    )
+    op.create_index(
+        "ix_file_form_evidence_grants_form_instance_id",
+        "file_form_evidence_grants",
+        ["form_instance_id"],
+    )
+    op.create_index(
+        "ix_file_form_evidence_grants_custodian_user_id",
+        "file_form_evidence_grants",
+        ["custodian_user_id"],
+    )
+    op.create_index(
+        "ix_file_form_evidence_grants_status",
+        "file_form_evidence_grants",
+        ["status"],
+    )
+    op.create_index(
+        "ix_file_form_evidence_grants_expires_at",
+        "file_form_evidence_grants",
+        ["expires_at"],
+    )
+    op.create_index(
+        "ix_file_form_evidence_grants_file_asset_id",
+        "file_form_evidence_grants",
+        ["file_asset_id"],
+    )
+    op.create_index(
+        "ix_file_form_evidence_grants_file_version_id",
+        "file_form_evidence_grants",
+        ["file_version_id"],
+    )
+    op.create_index(
+        "ix_file_form_evidence_grants_form",
+        "file_form_evidence_grants",
+        [
+            "tenant_id",
+            "form_instance_id",
+            "form_definition_id",
+            "form_definition_revision",
+        ],
+    )
+
+
+def downgrade() -> None:
+    op.drop_index(
+        "ix_file_form_evidence_grants_form",
+        table_name="file_form_evidence_grants",
+    )
+    op.drop_index(
+        "ix_file_form_evidence_grants_file_version_id",
+        table_name="file_form_evidence_grants",
+    )
+    op.drop_index(
+        "ix_file_form_evidence_grants_file_asset_id",
+        table_name="file_form_evidence_grants",
+    )
+    op.drop_index(
+        "ix_file_form_evidence_grants_expires_at",
+        table_name="file_form_evidence_grants",
+    )
+    op.drop_index(
+        "ix_file_form_evidence_grants_status",
+        table_name="file_form_evidence_grants",
+    )
+    op.drop_index(
+        "ix_file_form_evidence_grants_custodian_user_id",
+        table_name="file_form_evidence_grants",
+    )
+    op.drop_index(
+        "ix_file_form_evidence_grants_form_instance_id",
+        table_name="file_form_evidence_grants",
+    )
+    op.drop_index(
+        "ix_file_form_evidence_grants_tenant_id",
+        table_name="file_form_evidence_grants",
+    )
+    op.drop_table("file_form_evidence_grants")

src/govoplan_files/backend/migrations/versions/a2b3c4d5e6f9_file_asset_lifecycle.py

@@ -0,0 +1,52 @@
+"""govern managed file retention and legal holds
+
+Revision ID: a2b3c4d5e6f9
+Revises: f1a2b3c4d5e7
+"""
+
+from __future__ import annotations
+
+import sqlalchemy as sa
+from alembic import op
+
+
+revision = "a2b3c4d5e6f9"
+down_revision = "f1a2b3c4d5e7"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    with op.batch_alter_table("file_assets") as batch_op:
+        batch_op.add_column(
+            sa.Column("retained_until", sa.DateTime(timezone=True), nullable=True)
+        )
+        batch_op.add_column(
+            sa.Column(
+                "legal_hold", sa.Boolean(), nullable=False, server_default=sa.false()
+            )
+        )
+        batch_op.add_column(
+            sa.Column(
+                "lifecycle_revision", sa.Integer(), nullable=False, server_default="1"
+            )
+        )
+        batch_op.add_column(
+            sa.Column("lifecycle_reason", sa.String(length=500), nullable=True)
+        )
+        batch_op.create_index(
+            op.f("ix_file_assets_retained_until"), ["retained_until"], unique=False
+        )
+        batch_op.create_index(
+            op.f("ix_file_assets_legal_hold"), ["legal_hold"], unique=False
+        )
+
+
+def downgrade() -> None:
+    with op.batch_alter_table("file_assets") as batch_op:
+        batch_op.drop_index(op.f("ix_file_assets_legal_hold"))
+        batch_op.drop_index(op.f("ix_file_assets_retained_until"))
+        batch_op.drop_column("lifecycle_reason")
+        batch_op.drop_column("lifecycle_revision")
+        batch_op.drop_column("legal_hold")
+        batch_op.drop_column("retained_until")

src/govoplan_files/backend/migrations/versions/b8c9d0e1f2a4_file_share_lifecycle.py

@@ -0,0 +1,53 @@
+"""file share lifecycle
+
+Revision ID: b8c9d0e1f2a4
+Revises: a7b8c9d0e1f3
+Create Date: 2026-07-30 00:00:00.000000
+"""
+from __future__ import annotations
+
+from alembic import op
+import sqlalchemy as sa
+
+
+revision = "b8c9d0e1f2a4"
+down_revision = "a7b8c9d0e1f3"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    with op.batch_alter_table("file_shares") as batch_op:
+        batch_op.add_column(
+            sa.Column("expires_at", sa.DateTime(timezone=True), nullable=True)
+        )
+        batch_op.add_column(
+            sa.Column("revoked_by_user_id", sa.String(length=36), nullable=True)
+        )
+        batch_op.create_foreign_key(
+            op.f("fk_file_shares_revoked_by_user_id_access_users"),
+            "access_users",
+            ["revoked_by_user_id"],
+            ["id"],
+            ondelete="SET NULL",
+        )
+        batch_op.create_index(
+            op.f("ix_file_shares_expires_at"), ["expires_at"], unique=False
+        )
+        batch_op.create_index(
+            op.f("ix_file_shares_revoked_by_user_id"),
+            ["revoked_by_user_id"],
+            unique=False,
+        )
+
+
+def downgrade() -> None:
+    with op.batch_alter_table("file_shares") as batch_op:
+        batch_op.drop_index(op.f("ix_file_shares_revoked_by_user_id"))
+        batch_op.drop_index(op.f("ix_file_shares_expires_at"))
+        batch_op.drop_constraint(
+            op.f("fk_file_shares_revoked_by_user_id_access_users"),
+            type_="foreignkey",
+        )
+        batch_op.drop_column("revoked_by_user_id")
+        batch_op.drop_column("expires_at")

src/govoplan_files/backend/migrations/versions/c9d0e1f2a3b5_file_integrity_reconciliation.py

@@ -0,0 +1,168 @@
+"""file integrity reconciliation
+
+Revision ID: c9d0e1f2a3b5
+Revises: b8c9d0e1f2a4
+Create Date: 2026-07-30 00:00:00.000000
+"""
+from __future__ import annotations
+
+from alembic import op
+import sqlalchemy as sa
+
+
+revision = "c9d0e1f2a3b5"
+down_revision = "b8c9d0e1f2a4"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    with op.batch_alter_table("file_blobs") as batch_op:
+        batch_op.add_column(
+            sa.Column(
+                "integrity_status",
+                sa.String(length=30),
+                nullable=False,
+                server_default="unchecked",
+            )
+        )
+        batch_op.add_column(
+            sa.Column("integrity_checked_at", sa.DateTime(timezone=True), nullable=True)
+        )
+        batch_op.add_column(
+            sa.Column("integrity_failure", sa.String(length=100), nullable=True)
+        )
+        batch_op.add_column(
+            sa.Column("quarantined_at", sa.DateTime(timezone=True), nullable=True)
+        )
+        batch_op.create_index(
+            op.f("ix_file_blobs_integrity_status"),
+            ["integrity_status"],
+            unique=False,
+        )
+        batch_op.create_index(
+            op.f("ix_file_blobs_integrity_checked_at"),
+            ["integrity_checked_at"],
+            unique=False,
+        )
+        batch_op.create_index(
+            op.f("ix_file_blobs_quarantined_at"),
+            ["quarantined_at"],
+            unique=False,
+        )
+
+    op.create_table(
+        "file_integrity_scans",
+        sa.Column("id", sa.String(length=36), nullable=False),
+        sa.Column("tenant_id", sa.String(length=36), nullable=False),
+        sa.Column("storage_backend", sa.String(length=50), nullable=False),
+        sa.Column("storage_prefix", sa.String(length=1000), nullable=False),
+        sa.Column("status", sa.String(length=30), nullable=False),
+        sa.Column("phase", sa.String(length=30), nullable=False),
+        sa.Column("verify_checksums", sa.Boolean(), nullable=False),
+        sa.Column("batch_size", sa.Integer(), nullable=False),
+        sa.Column("blob_cursor", sa.String(length=36), nullable=True),
+        sa.Column("object_cursor", sa.String(length=1000), nullable=True),
+        sa.Column("scanned_blob_count", sa.Integer(), nullable=False),
+        sa.Column("verified_blob_count", sa.Integer(), nullable=False),
+        sa.Column("quarantined_blob_count", sa.Integer(), nullable=False),
+        sa.Column("scanned_object_count", sa.Integer(), nullable=False),
+        sa.Column("orphan_object_count", sa.Integer(), nullable=False),
+        sa.Column("created_by_user_id", sa.String(length=36), nullable=True),
+        sa.Column("started_at", sa.DateTime(timezone=True), nullable=True),
+        sa.Column("completed_at", sa.DateTime(timezone=True), nullable=True),
+        sa.Column("last_error", sa.String(length=255), nullable=True),
+        sa.Column("created_at", sa.DateTime(timezone=True), nullable=False),
+        sa.Column("updated_at", sa.DateTime(timezone=True), nullable=False),
+        sa.ForeignKeyConstraint(
+            ["created_by_user_id"],
+            ["access_users.id"],
+            name=op.f(
+                "fk_file_integrity_scans_created_by_user_id_access_users"
+            ),
+            ondelete="SET NULL",
+        ),
+        sa.PrimaryKeyConstraint("id", name=op.f("pk_file_integrity_scans")),
+    )
+    for column in ("tenant_id", "status", "created_by_user_id"):
+        op.create_index(
+            op.f(f"ix_file_integrity_scans_{column}"),
+            "file_integrity_scans",
+            [column],
+            unique=False,
+        )
+
+    op.create_table(
+        "file_integrity_findings",
+        sa.Column("id", sa.String(length=36), nullable=False),
+        sa.Column("scan_id", sa.String(length=36), nullable=False),
+        sa.Column("tenant_id", sa.String(length=36), nullable=False),
+        sa.Column("kind", sa.String(length=40), nullable=False),
+        sa.Column("state", sa.String(length=30), nullable=False),
+        sa.Column("blob_id", sa.String(length=36), nullable=True),
+        sa.Column("storage_key", sa.String(length=1000), nullable=False),
+        sa.Column("expected_size_bytes", sa.Integer(), nullable=True),
+        sa.Column("observed_size_bytes", sa.Integer(), nullable=True),
+        sa.Column("expected_checksum_sha256", sa.String(length=64), nullable=True),
+        sa.Column("observed_checksum_sha256", sa.String(length=64), nullable=True),
+        sa.Column("resolved_at", sa.DateTime(timezone=True), nullable=True),
+        sa.Column("resolved_by_user_id", sa.String(length=36), nullable=True),
+        sa.Column("created_at", sa.DateTime(timezone=True), nullable=False),
+        sa.Column("updated_at", sa.DateTime(timezone=True), nullable=False),
+        sa.ForeignKeyConstraint(
+            ["blob_id"],
+            ["file_blobs.id"],
+            name=op.f("fk_file_integrity_findings_blob_id_file_blobs"),
+            ondelete="SET NULL",
+        ),
+        sa.ForeignKeyConstraint(
+            ["resolved_by_user_id"],
+            ["access_users.id"],
+            name=op.f(
+                "fk_file_integrity_findings_resolved_by_user_id_access_users"
+            ),
+            ondelete="SET NULL",
+        ),
+        sa.ForeignKeyConstraint(
+            ["scan_id"],
+            ["file_integrity_scans.id"],
+            name=op.f(
+                "fk_file_integrity_findings_scan_id_file_integrity_scans"
+            ),
+            ondelete="CASCADE",
+        ),
+        sa.PrimaryKeyConstraint("id", name=op.f("pk_file_integrity_findings")),
+    )
+    for column in (
+        "scan_id",
+        "tenant_id",
+        "kind",
+        "state",
+        "blob_id",
+        "resolved_by_user_id",
+    ):
+        op.create_index(
+            op.f(f"ix_file_integrity_findings_{column}"),
+            "file_integrity_findings",
+            [column],
+            unique=False,
+        )
+    op.create_index(
+        "ix_file_integrity_findings_scan_state",
+        "file_integrity_findings",
+        ["scan_id", "state"],
+        unique=False,
+    )
+
+
+def downgrade() -> None:
+    op.drop_table("file_integrity_findings")
+    op.drop_table("file_integrity_scans")
+    with op.batch_alter_table("file_blobs") as batch_op:
+        batch_op.drop_index(op.f("ix_file_blobs_quarantined_at"))
+        batch_op.drop_index(op.f("ix_file_blobs_integrity_checked_at"))
+        batch_op.drop_index(op.f("ix_file_blobs_integrity_status"))
+        batch_op.drop_column("quarantined_at")
+        batch_op.drop_column("integrity_failure")
+        batch_op.drop_column("integrity_checked_at")
+        batch_op.drop_column("integrity_status")

src/govoplan_files/backend/migrations/versions/d0e1f2a3b4c6_file_content_protection.py

@@ -0,0 +1,79 @@
+"""file content-protection metadata
+
+Revision ID: d0e1f2a3b4c6
+Revises: c9d0e1f2a3b5
+Create Date: 2026-08-02 00:00:00.000000
+"""
+
+from __future__ import annotations
+
+from alembic import op
+import sqlalchemy as sa
+
+
+revision = "d0e1f2a3b4c6"
+down_revision = "c9d0e1f2a3b5"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    with op.batch_alter_table("file_blobs") as batch_op:
+        batch_op.add_column(
+            sa.Column(
+                "protection_discriminator",
+                sa.String(length=320),
+                nullable=False,
+                server_default="plaintext",
+            )
+        )
+        batch_op.add_column(
+            sa.Column("encryption_envelope_id", sa.String(length=255), nullable=True)
+        )
+        batch_op.add_column(
+            sa.Column("storage_checksum_sha256", sa.String(length=64), nullable=True)
+        )
+        batch_op.add_column(
+            sa.Column("storage_size_bytes", sa.Integer(), nullable=True)
+        )
+        batch_op.drop_constraint(
+            "uq_file_blobs_tenant_checksum_size",
+            type_="unique",
+        )
+        batch_op.create_unique_constraint(
+            "uq_file_blobs_tenant_checksum_size_protection",
+            [
+                "tenant_id",
+                "checksum_sha256",
+                "size_bytes",
+                "protection_discriminator",
+            ],
+        )
+        batch_op.create_index(
+            op.f("ix_file_blobs_protection_discriminator"),
+            ["protection_discriminator"],
+            unique=False,
+        )
+        batch_op.create_index(
+            op.f("ix_file_blobs_encryption_envelope_id"),
+            ["encryption_envelope_id"],
+            unique=False,
+        )
+
+
+def downgrade() -> None:
+    with op.batch_alter_table("file_blobs") as batch_op:
+        batch_op.drop_index(op.f("ix_file_blobs_encryption_envelope_id"))
+        batch_op.drop_index(op.f("ix_file_blobs_protection_discriminator"))
+        batch_op.drop_constraint(
+            "uq_file_blobs_tenant_checksum_size_protection",
+            type_="unique",
+        )
+        batch_op.create_unique_constraint(
+            "uq_file_blobs_tenant_checksum_size",
+            ["tenant_id", "checksum_sha256", "size_bytes"],
+        )
+        batch_op.drop_column("storage_size_bytes")
+        batch_op.drop_column("storage_checksum_sha256")
+        batch_op.drop_column("encryption_envelope_id")
+        batch_op.drop_column("protection_discriminator")

src/govoplan_files/backend/migrations/versions/f1a2b3c4d5e7_file_integrity_action_revisions.py

@@ -0,0 +1,34 @@
+"""add stale-action revisions to Files integrity operations
+
+Revision ID: f1a2b3c4d5e7
+Revises: d0e1f2a3b4c6
+"""
+
+from __future__ import annotations
+
+import sqlalchemy as sa
+from alembic import op
+
+
+revision = "f1a2b3c4d5e7"
+down_revision = "d0e1f2a3b4c6"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    with op.batch_alter_table("file_integrity_scans") as batch_op:
+        batch_op.add_column(
+            sa.Column("revision", sa.Integer(), nullable=False, server_default="1")
+        )
+    with op.batch_alter_table("file_integrity_findings") as batch_op:
+        batch_op.add_column(
+            sa.Column("revision", sa.Integer(), nullable=False, server_default="1")
+        )
+
+
+def downgrade() -> None:
+    with op.batch_alter_table("file_integrity_findings") as batch_op:
+        batch_op.drop_column("revision")
+    with op.batch_alter_table("file_integrity_scans") as batch_op:
+        batch_op.drop_column("revision")

src/govoplan_files/backend/operational_checks.py

@@ -0,0 +1,67 @@
+from __future__ import annotations
+
+import hashlib
+import secrets
+from uuid import uuid4
+
+from govoplan_core.core.operations import OperationalCheck
+from govoplan_files.backend.storage.backends import (
+    StorageBackendError,
+    get_storage_backend,
+)
+
+
+def managed_storage_roundtrip_check() -> OperationalCheck:
+    """Exercise the configured managed store without retaining probe data."""
+
+    backend = get_storage_backend()
+    key = f".govoplan-health/probes/{uuid4().hex}.bin"
+    payload = secrets.token_bytes(64)
+    expected_digest = hashlib.sha256(payload).hexdigest()
+    delete_error: Exception | None = None
+    try:
+        backend.put_bytes(key, payload, content_type="application/octet-stream")
+        stored = backend.get_bytes(key)
+        info = backend.stat(key)
+        if info.size_bytes != len(payload):
+            raise StorageBackendError("Managed storage returned an unexpected object size")
+        if hashlib.sha256(stored).hexdigest() != expected_digest:
+            raise StorageBackendError("Managed storage returned different bytes than were written")
+    except Exception as exc:  # noqa: BLE001 - operational boundary reports provider failures.
+        return OperationalCheck(
+            id="files.managed_storage_roundtrip",
+            label="Managed file storage",
+            state="error",
+            detail=(
+                "The configured managed file store failed a bounded write/read/stat/delete "
+                f"probe ({type(exc).__name__})."
+            ),
+            readiness_critical=True,
+            metrics={"backend": backend.name, "probe_bytes": len(payload)},
+        )
+    finally:
+        try:
+            backend.delete(key)
+        except Exception as exc:  # noqa: BLE001 - reported below when the data probe passed.
+            delete_error = exc
+
+    if delete_error is not None:
+        return OperationalCheck(
+            id="files.managed_storage_roundtrip",
+            label="Managed file storage",
+            state="error",
+            detail=(
+                "Managed file bytes round-tripped, but probe cleanup failed "
+                f"({type(delete_error).__name__})."
+            ),
+            readiness_critical=True,
+            metrics={"backend": backend.name, "probe_bytes": len(payload)},
+        )
+    return OperationalCheck(
+        id="files.managed_storage_roundtrip",
+        label="Managed file storage",
+        state="ok",
+        detail="The configured managed file store passed write, read, stat, integrity, and delete checks.",
+        metrics={"backend": backend.name, "probe_bytes": len(payload)},
+    )
+

src/govoplan_files/backend/provider_state.py

@@ -0,0 +1,133 @@
+from __future__ import annotations
+
+from collections import defaultdict
+from datetime import UTC, datetime
+
+from sqlalchemy import or_, select
+from sqlalchemy.orm import Session
+
+from govoplan_core.core.provider_governance import (
+    ExternalProviderRuntimeState,
+    ExternalProviderStateContext,
+)
+from govoplan_files.backend.db.models import FileConnectorProfile, FileConnectorSpace
+from govoplan_files.backend.storage.connector_providers import (
+    ConnectorProviderDescriptor,
+    connector_provider_descriptors,
+)
+
+
+REMOTE_STORAGE_PROVIDER_ID = "files.remote_storage"
+
+
+def remote_storage_provider_states(
+    context: ExternalProviderStateContext,
+) -> tuple[ExternalProviderRuntimeState, ...]:
+    if not isinstance(context.session, Session):
+        raise RuntimeError("Files provider state requires a database session.")
+    statement = select(FileConnectorProfile)
+    if context.tenant_id is not None:
+        statement = statement.where(
+            or_(
+                FileConnectorProfile.tenant_id.is_(None),
+                FileConnectorProfile.tenant_id == context.tenant_id,
+            )
+        )
+    profiles = tuple(
+        context.session.scalars(
+            statement.order_by(
+                FileConnectorProfile.tenant_id,
+                FileConnectorProfile.id,
+            ).limit(context.max_items + 1)
+        )
+    )
+    if not profiles:
+        return ()
+
+    profile_ids = tuple(item.id for item in profiles)
+    space_statement = select(FileConnectorSpace).where(
+        FileConnectorSpace.connector_profile_id.in_(profile_ids),
+        FileConnectorSpace.deleted_at.is_(None),
+    )
+    if context.tenant_id is not None:
+        space_statement = space_statement.where(
+            FileConnectorSpace.tenant_id == context.tenant_id
+        )
+    spaces_by_profile: dict[str, list[FileConnectorSpace]] = defaultdict(list)
+    for space in context.session.scalars(space_statement):
+        spaces_by_profile[space.connector_profile_id].append(space)
+
+    descriptors = {
+        item.provider: item for item in connector_provider_descriptors()
+    }
+    observed_at = datetime.now(UTC)
+    return tuple(
+        _profile_state(
+            profile,
+            spaces=spaces_by_profile.get(profile.id, []),
+            descriptor=descriptors.get(profile.provider),
+            observed_at=observed_at,
+        )
+        for profile in profiles
+    )
+
+
+def _profile_state(
+    profile: FileConnectorProfile,
+    *,
+    spaces: list[FileConnectorSpace],
+    descriptor: ConnectorProviderDescriptor | None,
+    observed_at: datetime,
+) -> ExternalProviderRuntimeState:
+    active_spaces = tuple(item for item in spaces if item.is_active)
+    active = bool(profile.enabled)
+    implementation_ready = bool(
+        descriptor is not None and descriptor.implemented and descriptor.installed
+    )
+    health = (
+        "inactive"
+        if not active
+        else "error"
+        if not implementation_ready
+        else "unknown"
+    )
+    recovery = (
+        "not_applicable"
+        if not active
+        else "unsupported"
+        if not implementation_ready
+        else "attention"
+    )
+    detail = (
+        "Remote-storage connector profile is disabled."
+        if not active
+        else "The configured provider is not available in this runtime."
+        if not implementation_ready
+        else "Software support is available; no live remote health observation is retained."
+    )
+    return ExternalProviderRuntimeState(
+        provider_id=REMOTE_STORAGE_PROVIDER_ID,
+        binding_ref=f"files:connector-profile:{profile.id}",
+        authority_mode="external_mirror",
+        observed_at=observed_at,
+        configured=True,
+        active=active,
+        health=health,
+        freshness="unknown" if active else "not_applicable",
+        conflict="not_applicable",
+        recovery=recovery,
+        detail=detail,
+        metrics={
+            "provider": profile.provider,
+            "configured_spaces": len(spaces),
+            "active_spaces": len(active_spaces),
+            "write_requested_spaces": sum(
+                1 for item in active_spaces if not item.read_only
+            ),
+            "software_implemented": bool(descriptor and descriptor.implemented),
+            "optional_dependency_available": bool(descriptor and descriptor.installed),
+        },
+    )
+
+
+__all__ = ["REMOTE_STORAGE_PROVIDER_ID", "remote_storage_provider_states"]

src/govoplan_files/backend/record_source.py

@@ -0,0 +1,113 @@
+from __future__ import annotations
+
+from collections.abc import Sequence
+from urllib.parse import quote
+
+from sqlalchemy.orm import Session
+
+from govoplan_core.core.records import (
+    RecordContractError,
+    RecordSourceLocator,
+    RecordSourceReference,
+)
+from govoplan_files.backend.db.models import FileBlob, FileVersion
+from govoplan_files.backend.storage.common import FileStorageError
+from govoplan_files.backend.storage.files import get_asset_for_user
+
+
+CAPABILITY_RECORD_SOURCE_FILES = "records.source.files"
+
+
+class FilesRecordSource:
+    provider_id = "files"
+
+    def resource_types(self) -> Sequence[str]:
+        return ("file_version",)
+
+    def resolve(
+        self,
+        session: object,
+        principal: object,
+        *,
+        locator: RecordSourceLocator,
+        purpose: str,
+    ) -> RecordSourceReference:
+        if not isinstance(session, Session):
+            raise RecordContractError(
+                "Files record references require a database session."
+            )
+        tenant_id = str(getattr(principal, "tenant_id", "") or "").strip()
+        if not tenant_id or locator.tenant_id != tenant_id:
+            raise RecordContractError("Files record references cannot cross tenants.")
+        if locator.source_module != "files" or locator.resource_type != "file_version":
+            raise RecordContractError("Unsupported Files record source type.")
+        if not str(purpose or "").strip():
+            raise RecordContractError("Files record references require a purpose.")
+        if not hasattr(principal, "has") or not (
+            principal.has("files:file:read") or principal.has("files:file:admin")
+        ):
+            raise RecordContractError("Current Files read permission is required.")
+        user = getattr(principal, "user", None)
+        user_id = str(
+            getattr(user, "id", "") or getattr(principal, "membership_id", "") or ""
+        ).strip()
+        if not user_id:
+            raise RecordContractError(
+                "Files record references require a tenant user principal."
+            )
+        try:
+            asset = get_asset_for_user(
+                session,
+                tenant_id=tenant_id,
+                user_id=user_id,
+                asset_id=locator.resource_id,
+                is_admin=principal.has("files:file:admin"),
+            )
+        except FileStorageError as exc:
+            raise RecordContractError(str(exc)) from exc
+        version_query = session.query(FileVersion).filter(
+            FileVersion.tenant_id == tenant_id,
+            FileVersion.file_asset_id == asset.id,
+        )
+        revision = locator.source_revision.strip()
+        version = version_query.filter(FileVersion.id == revision).one_or_none()
+        if version is None:
+            raise RecordContractError("The exact file version does not exist.")
+        blob = session.get(FileBlob, version.blob_id)
+        if blob is None or blob.tenant_id != tenant_id:
+            raise RecordContractError(
+                "The exact file version has no managed content object."
+            )
+        if blob.quarantined_at is not None or blob.integrity_status == "failed":
+            raise RecordContractError(
+                "The exact file version failed the current integrity gate."
+            )
+        return RecordSourceReference(
+            locator=locator,
+            label=version.filename_at_upload,
+            authority_mode="external_authoritative",
+            content_sha256=version.checksum_sha256,
+            content_type=version.content_type,
+            size_bytes=version.size_bytes,
+            recorded_at=version.created_at,
+            launch_url=(
+                f"/files?fileId={quote(asset.id, safe='')}&versionId={quote(version.id, safe='')}"
+            ),
+            metadata={
+                "display_path": version.display_path_at_upload,
+                "version_number": version.version_number,
+                "integrity_status": blob.integrity_status,
+                "protection": blob.protection_discriminator,
+            },
+        )
+
+
+def create_files_record_source(_context: object) -> FilesRecordSource:
+    return FilesRecordSource()
+
+
+__all__ = [
+    "CAPABILITY_RECORD_SOURCE_FILES",
+    "FilesRecordSource",
+    "create_files_record_source",
+]

src/govoplan_files/backend/routes/__init__.py

@@ -0,0 +1 @@
+"""Focused HTTP route modules for the Files API."""

src/govoplan_files/backend/routes/assets.py

@@ -0,0 +1,175 @@
+from __future__ import annotations
+
+from io import BytesIO
+from fastapi import APIRouter, Depends
+from fastapi.responses import StreamingResponse
+from sqlalchemy.orm import Session
+
+from govoplan_core.auth import ApiPrincipal, require_scope
+from govoplan_files.backend.schemas import (
+    BulkDeleteRequest,
+    BulkDeleteResponse,
+    FileAssetResponse,
+)
+from govoplan_core.db.session import get_session
+from govoplan_files.backend.storage.common import FileStorageError
+from govoplan_files.backend.storage.files import (
+    get_asset_for_user,
+    read_asset_bytes,
+    read_asset_version_bytes,
+    soft_delete_assets,
+)
+
+
+from govoplan_files.backend.route_support import (
+    _asset_response,
+    _attachment_disposition,
+    _audit_connector_event,
+    _http_error,
+    _is_admin,
+)
+
+router = APIRouter(prefix="/files", tags=["files"])
+
+
+@router.get("/{file_id}", response_model=FileAssetResponse)
+def get_file(
+    file_id: str,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:read")),
+):
+    try:
+        asset = get_asset_for_user(
+            session,
+            tenant_id=principal.tenant_id,
+            user_id=principal.user.id,
+            asset_id=file_id,
+            is_admin=_is_admin(principal),
+        )
+        return _asset_response(session, asset, include_shares=True)
+    except FileStorageError as exc:
+        raise _http_error(exc, not_found=True) from exc
+
+
+@router.get("/{file_id}/download")
+def download_file(
+    file_id: str,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:download")),
+):
+    try:
+        asset = get_asset_for_user(
+            session,
+            tenant_id=principal.tenant_id,
+            user_id=principal.user.id,
+            asset_id=file_id,
+            is_admin=_is_admin(principal),
+        )
+        data, version, blob = read_asset_bytes(session, asset)
+        _audit_connector_event(
+            session,
+            principal,
+            action="files.connector.accessed",
+            asset=asset,
+            version=version,
+            blob=blob,
+            operation="download",
+            commit=True,
+        )
+    except FileStorageError as exc:
+        raise _http_error(exc, not_found=True) from exc
+    headers = {"Content-Disposition": _attachment_disposition(asset.filename)}
+    return StreamingResponse(
+        BytesIO(data),
+        media_type=blob.content_type or "application/octet-stream",
+        headers=headers,
+    )
+
+
+@router.get("/{file_id}/versions/{version_id}/download")
+def download_file_version(
+    file_id: str,
+    version_id: str,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:download")),
+):
+    try:
+        asset = get_asset_for_user(
+            session,
+            tenant_id=principal.tenant_id,
+            user_id=principal.user.id,
+            asset_id=file_id,
+            is_admin=_is_admin(principal),
+        )
+        data, version, blob = read_asset_version_bytes(session, asset, version_id)
+        _audit_connector_event(
+            session,
+            principal,
+            action="files.connector.accessed",
+            asset=asset,
+            version=version,
+            blob=blob,
+            operation="download-version",
+            commit=True,
+        )
+    except FileStorageError as exc:
+        raise _http_error(exc, not_found=True) from exc
+    headers = {
+        "Content-Disposition": _attachment_disposition(
+            version.filename_at_upload or asset.filename
+        )
+    }
+    return StreamingResponse(
+        BytesIO(data),
+        media_type=blob.content_type or "application/octet-stream",
+        headers=headers,
+    )
+
+
+@router.delete("/{file_id}", response_model=BulkDeleteResponse)
+def delete_file(
+    file_id: str,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:delete")),
+):
+    try:
+        asset = get_asset_for_user(
+            session,
+            tenant_id=principal.tenant_id,
+            user_id=principal.user.id,
+            asset_id=file_id,
+            require_write=True,
+            is_admin=_is_admin(principal),
+        )
+        count = soft_delete_assets(session, [asset])
+        session.commit()
+        return BulkDeleteResponse(deleted_count=count)
+    except FileStorageError as exc:
+        session.rollback()
+        raise _http_error(exc, not_found=True) from exc
+
+
+@router.post("/bulk-delete", response_model=BulkDeleteResponse)
+def bulk_delete_files(
+    payload: BulkDeleteRequest,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:delete")),
+):
+    try:
+        assets = [
+            get_asset_for_user(
+                session,
+                tenant_id=principal.tenant_id,
+                user_id=principal.user.id,
+                asset_id=file_id,
+                require_write=True,
+                is_admin=_is_admin(principal),
+            )
+            for file_id in payload.file_ids
+        ]
+        count = soft_delete_assets(session, assets)
+        session.commit()
+        return BulkDeleteResponse(deleted_count=count)
+    except FileStorageError as exc:
+        session.rollback()
+        raise _http_error(exc) from exc

src/govoplan_files/backend/routes/connector_io.py

@@ -0,0 +1,631 @@
+from __future__ import annotations
+
+import json
+from fastapi import APIRouter, Depends, HTTPException, status
+from sqlalchemy.orm import Session
+
+from govoplan_core.auth import ApiPrincipal, require_any_scope, require_scope
+from govoplan_core.audit.logging import audit_from_principal
+from govoplan_files.backend.schemas import (
+    FileConnectorBrowseItem,
+    FileConnectorBrowseResponse,
+    FileConnectorFolderSyncItemResponse,
+    FileConnectorFolderSyncRequest,
+    FileConnectorFolderSyncResponse,
+    FileConnectorFolderSyncSummary,
+    FileConnectorImportRequest,
+    FileConnectorSyncResponse,
+    FileConnectorWriteRequest,
+    FileConnectorWriteResponse,
+    FileUploadResponse,
+)
+from govoplan_core.db.session import get_session
+from govoplan_files.backend.storage.paths import UnsafeFilePathError
+from govoplan_files.backend.storage.common import FileStorageError
+from govoplan_files.backend.storage.connector_browse import (
+    ConnectorBrowseError,
+    ConnectorBrowseUnsupported,
+    browse_connector_profile,
+    normalize_connector_browse_path,
+)
+from govoplan_files.backend.storage.connector_imports import (
+    ConnectorImportError,
+    ConnectorImportUnsupported,
+)
+from govoplan_files.backend.storage.connector_folder_sync import (
+    connector_relative_path,
+    discover_connector_folder,
+    join_connector_path,
+)
+from govoplan_files.backend.storage.connector_deployment import (
+    connector_effective_endpoint_url,
+)
+from govoplan_files.backend.storage.connector_policy import (
+    ConnectorAccessRequest,
+    ConnectorPolicyDenied,
+    connector_policy_decision,
+)
+from govoplan_files.backend.storage.files import (
+    create_file_asset,
+    get_asset_for_user,
+    read_asset_bytes,
+    sync_file_asset_from_source,
+)
+from govoplan_files.backend.storage.connector_spaces import (
+    connector_space_owner_id,
+    get_connector_space_for_user,
+)
+from govoplan_files.backend.storage.connector_writes import write_connector_file
+
+
+from govoplan_files.backend.route_support import (
+    _asset_response,
+    _audit_connector_imports,
+    _audit_connector_sync,
+    _connector_browse_next_token,
+    _connector_policy_error,
+    _connector_space_policy_decision,
+    _download_connector_payload,
+    _ensure_campaign_file_access,
+    _http_error,
+    _is_admin,
+    _visible_connector_profile,
+)
+
+router = APIRouter(prefix="/files", tags=["files"])
+
+
+@router.post(
+    "/connector-spaces/{space_id}/sync",
+    response_model=FileConnectorFolderSyncResponse,
+)
+def sync_connector_space_folder(
+    space_id: str,
+    payload: FileConnectorFolderSyncRequest,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:upload")),
+):
+    try:
+        space = get_connector_space_for_user(
+            session,
+            tenant_id=principal.tenant_id,
+            user_id=principal.user.id,
+            space_id=space_id,
+            is_admin=_is_admin(principal),
+        )
+    except FileStorageError as exc:
+        session.rollback()
+        raise _http_error(exc, not_found=True) from exc
+
+    try:
+        if space.sync_mode != "manual":
+            raise FileStorageError("This connector space is not configured for manual sync")
+        profile = _visible_connector_profile(
+            session, principal, space.connector_profile_id
+        )
+        remote_path = join_connector_path(space.remote_path, payload.path)
+        target_folder = join_connector_path(payload.target_folder)
+        decision = _connector_space_policy_decision(
+            profile,
+            library_id=space.library_id,
+            remote_path=remote_path,
+            operation="sync",
+        )
+        if not decision.allowed:
+            raise ConnectorPolicyDenied(decision)
+        discovery = discover_connector_folder(
+            profile,
+            path=remote_path,
+            library_id=space.library_id,
+            recursive=payload.recursive,
+            max_files=payload.max_files,
+            max_depth=payload.max_depth,
+        )
+    except ConnectorPolicyDenied as exc:
+        session.rollback()
+        raise _connector_policy_error(exc) from exc
+    except ConnectorBrowseUnsupported as exc:
+        session.rollback()
+        raise HTTPException(
+            status_code=status.HTTP_501_NOT_IMPLEMENTED, detail=str(exc)
+        ) from exc
+    except (ConnectorBrowseError, FileStorageError, ValueError) as exc:
+        session.rollback()
+        raise _http_error(exc) from exc
+
+    owner_id = connector_space_owner_id(space)
+    items: list[FileConnectorFolderSyncItemResponse] = [
+        FileConnectorFolderSyncItemResponse(
+            source_path=skipped.path,
+            action="skipped",
+            detail=skipped.reason,
+        )
+        for skipped in discovery.skipped
+    ]
+    counts = {
+        "created": 0,
+        "updated": 0,
+        "unchanged": 0,
+        "skipped": len(discovery.skipped),
+        "conflicts": 0,
+        "policy_denied": 0,
+        "failed": 0,
+    }
+
+    for source in discovery.files:
+        source_path = normalize_connector_browse_path(source.path)
+        try:
+            relative_path = connector_relative_path(
+                space_root=space.remote_path,
+                item_path=source_path,
+            )
+            target_path = join_connector_path(target_folder, relative_path)
+        except (ConnectorBrowseError, ValueError) as exc:
+            counts["failed"] += 1
+            items.append(
+                FileConnectorFolderSyncItemResponse(
+                    source_path=source_path,
+                    action="failed",
+                    detail=str(exc),
+                )
+            )
+            continue
+
+        try:
+            with session.begin_nested():
+                file_payload = FileConnectorImportRequest(
+                    library_id=space.library_id or "",
+                    path=source_path,
+                    owner_type=space.owner_type,
+                    owner_id=owner_id,
+                    target_path=target_path,
+                    conflict_strategy=payload.conflict_strategy,
+                    source_revision=source.etag,
+                    metadata={
+                        **dict(source.metadata),
+                        **payload.metadata,
+                        "connector_space_id": space.id,
+                        "browse_name": source.name,
+                        "browse_path": source_path,
+                        "browse_modified_at": source.modified_at,
+                        "browse_etag": source.etag,
+                        "folder_sync": True,
+                    },
+                )
+                _source_path, downloaded, metadata = _download_connector_payload(
+                    profile, file_payload, operation="sync"
+                )
+                stored, sync_action, previous_version_id = (
+                    sync_file_asset_from_source(
+                        session,
+                        tenant_id=principal.tenant_id,
+                        owner_type=space.owner_type,
+                        owner_id=owner_id,
+                        user_id=principal.user.id,
+                        filename=downloaded.filename,
+                        data=downloaded.data,
+                        display_path=target_path,
+                        content_type=downloaded.content_type,
+                        metadata=metadata,
+                        conflict_strategy=payload.conflict_strategy,
+                        is_admin=_is_admin(principal),
+                    )
+                )
+                _audit_connector_sync(
+                    session,
+                    principal,
+                    stored.asset,
+                    sync_action=sync_action,
+                    previous_version_id=previous_version_id,
+                )
+                file_response = _asset_response(
+                    session, stored.asset, include_shares=True
+                )
+            counts[sync_action] += 1
+            items.append(
+                FileConnectorFolderSyncItemResponse(
+                    source_path=source_path,
+                    target_path=target_path,
+                    action=sync_action,
+                    file=file_response,
+                    previous_version_id=previous_version_id,
+                    current_version_id=stored.version.id,
+                    source_revision=file_response.source_revision,
+                )
+            )
+        except ConnectorPolicyDenied as exc:
+            counts["policy_denied"] += 1
+            items.append(
+                FileConnectorFolderSyncItemResponse(
+                    source_path=source_path,
+                    target_path=target_path,
+                    action="policy_denied",
+                    source_revision=source.etag,
+                    detail=str(exc),
+                    policy_decision=exc.decision.to_dict(),
+                )
+            )
+        except FileStorageError as exc:
+            detail = str(exc)
+            if detail.startswith("Skipped upload target:"):
+                action = "skipped"
+                counts["skipped"] += 1
+            elif detail.startswith("Target file already exists:"):
+                action = "conflict"
+                counts["conflicts"] += 1
+            else:
+                action = "failed"
+                counts["failed"] += 1
+            items.append(
+                FileConnectorFolderSyncItemResponse(
+                    source_path=source_path,
+                    target_path=target_path,
+                    action=action,
+                    source_revision=source.etag,
+                    detail=detail,
+                )
+            )
+        except (
+            ConnectorImportError,
+            UnsafeFilePathError,
+            OSError,
+            ValueError,
+            json.JSONDecodeError,
+        ) as exc:
+            counts["failed"] += 1
+            items.append(
+                FileConnectorFolderSyncItemResponse(
+                    source_path=source_path,
+                    target_path=target_path,
+                    action="failed",
+                    source_revision=source.etag,
+                    detail=str(exc),
+                )
+            )
+
+    summary = FileConnectorFolderSyncSummary(
+        discovered=len(discovery.files),
+        **counts,
+    )
+    audit_from_principal(
+        session,
+        principal,
+        action="files.connector.folder_synced",
+        object_type="file_connector_space",
+        object_id=space.id,
+        details={
+            "connector_profile_id": profile.id,
+            "provider": profile.provider,
+            "library_id": space.library_id,
+            "remote_path": remote_path,
+            "target_folder": target_folder,
+            "recursive": payload.recursive,
+            "truncated": discovery.truncated,
+            "summary": summary.model_dump(),
+            "results": [
+                {
+                    "source_path": item.source_path,
+                    "target_path": item.target_path,
+                    "action": item.action,
+                    "file_id": item.file.id if item.file else None,
+                    "current_version_id": item.current_version_id,
+                }
+                for item in items
+            ],
+        },
+    )
+    session.commit()
+    return FileConnectorFolderSyncResponse(
+        connector_space_id=space.id,
+        connector_profile_id=profile.id,
+        provider=profile.provider,
+        remote_path=remote_path,
+        target_folder=target_folder,
+        recursive=payload.recursive,
+        truncated=discovery.truncated,
+        summary=summary,
+        items=items,
+    )
+
+
+@router.post(
+    "/connector-spaces/{space_id}/write-back",
+    response_model=FileConnectorWriteResponse,
+)
+def write_back_connector_file(
+    space_id: str,
+    payload: FileConnectorWriteRequest,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:connector:write")),
+):
+    try:
+        space = get_connector_space_for_user(
+            session,
+            tenant_id=principal.tenant_id,
+            user_id=principal.user.id,
+            space_id=space_id,
+            is_admin=_is_admin(principal),
+        )
+        if space.read_only:
+            raise FileStorageError("This connector space is read-only")
+        profile = _visible_connector_profile(
+            session, principal, space.connector_profile_id
+        )
+        requested_path = normalize_connector_browse_path(payload.remote_path)
+        remote_path = normalize_connector_browse_path(
+            "/".join(part for part in (space.remote_path, requested_path) if part)
+        )
+        decision = connector_policy_decision(
+            ConnectorAccessRequest(
+                connector_id=profile.id,
+                credential_id=profile.credential_profile_id,
+                provider=profile.provider,
+                external_path=remote_path,
+                external_url=connector_effective_endpoint_url(
+                    provider=profile.provider,
+                    endpoint_url=profile.endpoint_url,
+                    metadata=profile.metadata,
+                ),
+                operation="write",
+            ),
+            profile.policy_sources,
+        )
+        if not decision.allowed:
+            raise ConnectorPolicyDenied(decision)
+        asset = get_asset_for_user(
+            session,
+            tenant_id=principal.tenant_id,
+            user_id=principal.user.id,
+            asset_id=payload.file_id,
+            require_write=True,
+            is_admin=_is_admin(principal),
+        )
+        data, version, blob = read_asset_bytes(session, asset)
+        connector_library_id = space.library_id
+        content_type = blob.content_type
+        # Close the read snapshot before the independent recovery transaction
+        # records authority for the external effect. This avoids upgrading an
+        # older SQLite read snapshot after the ledger commit.
+        session.commit()
+        result = write_connector_file(
+            profile,
+            tenant_id=principal.tenant_id,
+            library_id=connector_library_id,
+            remote_path=remote_path,
+            data=data,
+            content_type=content_type,
+            expected_revision=payload.expected_revision,
+            idempotency_key=payload.idempotency_key,
+        )
+        audit_from_principal(
+            session,
+            principal,
+            action="files.connector.written",
+            object_type="file",
+            object_id=asset.id,
+            details={
+                "file_version_id": version.id,
+                "file_blob_id": blob.id,
+                "checksum_sha256": blob.checksum_sha256,
+                "connector_space_id": space.id,
+                "connector_profile_id": profile.id,
+                "remote_path": remote_path,
+                "recovery_operation_id": result.recovery_operation_id,
+                "recovery_status": result.status,
+                "revision": result.revision,
+            },
+        )
+        session.commit()
+        return FileConnectorWriteResponse(
+            recovery_operation_id=result.recovery_operation_id,
+            status=result.status,
+            replayed=result.replayed,
+            provider=result.provider,
+            remote_path=result.remote_path,
+            revision=result.revision,
+            checksum_sha256=result.checksum_sha256,
+            size_bytes=result.size_bytes,
+        )
+    except ConnectorPolicyDenied as exc:
+        session.rollback()
+        raise _connector_policy_error(exc) from exc
+    except (FileStorageError, ConnectorBrowseError) as exc:
+        session.rollback()
+        raise _http_error(exc) from exc
+
+
+@router.post(
+    "/connectors/profiles/{profile_id}/import", response_model=FileUploadResponse
+)
+def import_connector_file(
+    profile_id: str,
+    payload: FileConnectorImportRequest,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:upload")),
+):
+    try:
+        if payload.campaign_id:
+            _ensure_campaign_file_access(session, principal, payload.campaign_id)
+        profile = _visible_connector_profile(
+            session, principal, profile_id, campaign_id=payload.campaign_id
+        )
+        _source_path, downloaded, metadata = _download_connector_payload(
+            profile, payload, operation="import"
+        )
+        target_owner = payload.owner_id or principal.user.id
+        stored = create_file_asset(
+            session,
+            tenant_id=principal.tenant_id,
+            owner_type=payload.owner_type,
+            owner_id=target_owner,
+            user_id=principal.user.id,
+            filename=downloaded.filename,
+            data=downloaded.data,
+            folder=payload.target_folder,
+            display_path=payload.target_path,
+            content_type=downloaded.content_type,
+            metadata=metadata,
+            campaign_id=payload.campaign_id,
+            conflict_strategy=payload.conflict_strategy,
+            is_admin=_is_admin(principal),
+        )
+        _audit_connector_imports(session, principal, [stored.asset])
+        session.commit()
+    except ConnectorPolicyDenied as exc:
+        session.rollback()
+        raise _connector_policy_error(exc) from exc
+    except ConnectorImportUnsupported as exc:
+        session.rollback()
+        raise HTTPException(
+            status_code=status.HTTP_501_NOT_IMPLEMENTED, detail=str(exc)
+        ) from exc
+    except (
+        ConnectorImportError,
+        FileStorageError,
+        UnsafeFilePathError,
+        ValueError,
+        json.JSONDecodeError,
+    ) as exc:
+        session.rollback()
+        raise _http_error(exc) from exc
+    return FileUploadResponse(
+        files=[_asset_response(session, stored.asset, include_shares=True)]
+    )
+
+
+@router.post(
+    "/connectors/profiles/{profile_id}/sync", response_model=FileConnectorSyncResponse
+)
+def sync_connector_file(
+    profile_id: str,
+    payload: FileConnectorImportRequest,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:upload")),
+):
+    try:
+        if payload.campaign_id:
+            _ensure_campaign_file_access(session, principal, payload.campaign_id)
+        profile = _visible_connector_profile(
+            session, principal, profile_id, campaign_id=payload.campaign_id
+        )
+        _source_path, downloaded, metadata = _download_connector_payload(
+            profile, payload, operation="sync"
+        )
+        target_owner = payload.owner_id or principal.user.id
+        stored, sync_action, previous_version_id = sync_file_asset_from_source(
+            session,
+            tenant_id=principal.tenant_id,
+            owner_type=payload.owner_type,
+            owner_id=target_owner,
+            user_id=principal.user.id,
+            filename=downloaded.filename,
+            data=downloaded.data,
+            folder=payload.target_folder,
+            display_path=payload.target_path,
+            content_type=downloaded.content_type,
+            metadata=metadata,
+            campaign_id=payload.campaign_id,
+            conflict_strategy=payload.conflict_strategy,
+            is_admin=_is_admin(principal),
+        )
+        _audit_connector_sync(
+            session,
+            principal,
+            stored.asset,
+            sync_action=sync_action,
+            previous_version_id=previous_version_id,
+        )
+        session.commit()
+    except ConnectorPolicyDenied as exc:
+        session.rollback()
+        raise _connector_policy_error(exc) from exc
+    except ConnectorImportUnsupported as exc:
+        session.rollback()
+        raise HTTPException(
+            status_code=status.HTTP_501_NOT_IMPLEMENTED, detail=str(exc)
+        ) from exc
+    except (
+        ConnectorImportError,
+        FileStorageError,
+        UnsafeFilePathError,
+        ValueError,
+        json.JSONDecodeError,
+    ) as exc:
+        session.rollback()
+        raise _http_error(exc) from exc
+    return FileConnectorSyncResponse(
+        file=_asset_response(session, stored.asset, include_shares=True),
+        action=sync_action,
+        previous_version_id=previous_version_id,
+        current_version_id=stored.version.id,
+    )
+
+
+@router.get(
+    "/connectors/profiles/{profile_id}/browse",
+    response_model=FileConnectorBrowseResponse,
+)
+def browse_connector_profile_items(
+    profile_id: str,
+    path: str | None = None,
+    library_id: str | None = None,
+    continuation_token: str | None = None,
+    campaign_id: str | None = None,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(
+        require_any_scope(
+            "files:file:read",
+            "files:file:upload",
+            "files:file:download",
+            "files:file:admin",
+            "system:settings:read",
+            "admin:settings:read",
+        )
+    ),
+):
+    try:
+        profile = _visible_connector_profile(
+            session, principal, profile_id, campaign_id=campaign_id
+        )
+        browse_path = normalize_connector_browse_path(path)
+        decision = connector_policy_decision(
+            ConnectorAccessRequest(
+                connector_id=profile.id,
+                credential_id=profile.credential_profile_id,
+                provider=profile.provider,
+                external_path=browse_path,
+                external_url=connector_effective_endpoint_url(
+                    provider=profile.provider,
+                    endpoint_url=profile.endpoint_url,
+                    metadata=profile.metadata,
+                ),
+                operation="browse",
+            ),
+            profile.policy_sources,
+        )
+        if not decision.allowed:
+            raise ConnectorPolicyDenied(decision)
+        items = browse_connector_profile(
+            profile,
+            path=browse_path,
+            library_id=library_id,
+            continuation_token=continuation_token,
+        )
+    except ConnectorPolicyDenied as exc:
+        raise _connector_policy_error(exc) from exc
+    except ConnectorBrowseUnsupported as exc:
+        raise HTTPException(
+            status_code=status.HTTP_501_NOT_IMPLEMENTED, detail=str(exc)
+        ) from exc
+    except (ConnectorBrowseError, OSError, ValueError, json.JSONDecodeError) as exc:
+        raise _http_error(exc) from exc
+    return FileConnectorBrowseResponse(
+        profile_id=profile.id,
+        provider=profile.provider,
+        path=browse_path,
+        library_id=library_id,
+        next_continuation_token=_connector_browse_next_token(items),
+        has_more=any(bool(item.metadata.get("listing_truncated")) for item in items),
+        decision=decision.to_dict(),
+        items=[FileConnectorBrowseItem(**item.to_response()) for item in items],
+    )

src/govoplan_files/backend/routes/connector_profiles.py

@@ -0,0 +1,261 @@
+from __future__ import annotations
+
+import json
+from fastapi import APIRouter, Depends
+from sqlalchemy.orm import Session
+
+from govoplan_core.auth import ApiPrincipal, require_any_scope
+from govoplan_files.backend.change_tracking import (
+    FILES_CONNECTOR_PROFILES_COLLECTION,
+)
+from govoplan_files.backend.schemas import (
+    FileConnectorProfileResponse,
+    FileConnectorProfileUpdateRequest,
+)
+from govoplan_core.db.session import get_session
+from govoplan_files.backend.storage.common import FileStorageError
+from govoplan_files.backend.storage.connector_credential_deletion import (
+    delete_connector_profile_row,
+)
+from govoplan_files.backend.storage.connector_deployment import (
+    connector_effective_endpoint_url,
+)
+from govoplan_files.backend.storage.connector_profile_store import (
+    connector_profile_from_row,
+    get_connector_profile_row,
+    update_connector_profile_row,
+)
+from govoplan_files.backend.storage.connector_policy import (
+    ConnectorPolicyDenied,
+)
+
+
+from govoplan_files.backend.route_support import (
+    FILES_CONNECTOR_PROFILE_RESOURCE,
+    _can_read_disabled_connector_profiles,
+    _connector_policy_error,
+    _credential_row_for_profile,
+    _ensure_connector_configuration_allowed,
+    _ensure_connector_local_policy_allowed,
+    _http_error,
+    _record_connector_settings_change,
+    _require_connector_profile_write,
+    _visible_connector_profile,
+)
+
+router = APIRouter(prefix="/files", tags=["files"])
+
+
+@router.get(
+    "/connectors/profiles/{profile_id}", response_model=FileConnectorProfileResponse
+)
+def get_connector_profile(
+    profile_id: str,
+    campaign_id: str | None = None,
+    include_disabled: bool = False,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(
+        require_any_scope(
+            "files:file:read",
+            "files:file:upload",
+            "files:file:download",
+            "files:file:admin",
+            "system:settings:read",
+            "admin:settings:read",
+        )
+    ),
+):
+    try:
+        profile = _visible_connector_profile(
+            session,
+            principal,
+            profile_id,
+            campaign_id=campaign_id,
+            include_disabled=include_disabled
+            and _can_read_disabled_connector_profiles(principal),
+            include_effective_policy=False,
+        )
+    except (OSError, ValueError, json.JSONDecodeError) as exc:
+        raise _http_error(exc) from exc
+    return FileConnectorProfileResponse(**profile.to_response())
+
+
+@router.patch(
+    "/connectors/profiles/{profile_id}", response_model=FileConnectorProfileResponse
+)
+def update_connector_profile(
+    profile_id: str,
+    payload: FileConnectorProfileUpdateRequest,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(
+        require_any_scope(
+            "files:file:admin", "system:settings:write", "admin:settings:write"
+        )
+    ),
+):
+    try:
+        row = get_connector_profile_row(
+            session,
+            tenant_id=principal.tenant_id,
+            profile_id=profile_id,
+            include_disabled=True,
+        )
+    except FileStorageError as exc:
+        raise _http_error(exc, not_found=True) from exc
+    _require_connector_profile_write(principal, row.scope_type)
+    credentials = payload.credentials
+    try:
+        credential_profile_id = (
+            payload.credential_profile_id
+            if payload.credential_profile_id is not None
+            else row.credential_profile_id
+        )
+        provider = payload.provider if payload.provider is not None else row.provider
+        credential_row = _credential_row_for_profile(
+            session,
+            principal,
+            credential_profile_id=credential_profile_id,
+            provider=provider,
+            profile_id=row.id,
+            scope_type=row.scope_type,
+            scope_id=row.scope_id,
+            include_disabled=True,
+        )
+        _ensure_connector_configuration_allowed(
+            session,
+            principal,
+            connector_id=row.id,
+            credential_id=credential_profile_id,
+            provider=provider,
+            endpoint_url=connector_effective_endpoint_url(
+                provider=provider,
+                endpoint_url=payload.endpoint_url
+                if payload.endpoint_url is not None
+                else row.endpoint_url,
+                metadata=payload.metadata
+                if payload.metadata is not None
+                else row.metadata_,
+            ),
+            base_path=payload.base_path
+            if payload.base_path is not None
+            else row.base_path,
+            scope_type=row.scope_type,
+            scope_id=row.scope_id,
+            operation="configure",
+        )
+        if payload.policy is not None:
+            _ensure_connector_local_policy_allowed(
+                session,
+                principal,
+                scope_type=row.scope_type,
+                scope_id=row.scope_id,
+                policy=payload.policy,
+            )
+        update_connector_profile_row(
+            session,
+            row,
+            user_id=principal.user.id,
+            label=payload.label,
+            provider=payload.provider,
+            endpoint_url=payload.endpoint_url,
+            base_path=payload.base_path,
+            enabled=payload.enabled,
+            credential_profile_id=payload.credential_profile_id,
+            credential_mode=payload.credential_mode,
+            username=credentials.username if credentials else None,
+            password=credentials.password if credentials else None,
+            token=credentials.token if credentials else None,
+            password_env=credentials.password_env if credentials else None,
+            token_env=credentials.token_env if credentials else None,
+            secret_ref=credentials.secret_ref if credentials else None,
+            clear_password=payload.clear_password,
+            clear_token=payload.clear_token,
+            capabilities=payload.capabilities,
+            policy=payload.policy,
+            metadata=payload.metadata,
+        )
+        _record_connector_settings_change(
+            session,
+            collection=FILES_CONNECTOR_PROFILES_COLLECTION,
+            resource_type=FILES_CONNECTOR_PROFILE_RESOURCE,
+            resource_id=row.id,
+            operation="updated",
+            principal=principal,
+            tenant_id=row.tenant_id,
+            payload={
+                "scope_type": row.scope_type,
+                "scope_id": row.scope_id,
+                "provider": row.provider,
+            },
+        )
+        session.commit()
+        session.refresh(row)
+        return FileConnectorProfileResponse(
+            **connector_profile_from_row(
+                row, credential_row=credential_row
+            ).to_response()
+        )
+    except ConnectorPolicyDenied as exc:
+        session.rollback()
+        raise _connector_policy_error(exc) from exc
+    except (FileStorageError, ValueError, json.JSONDecodeError) as exc:
+        session.rollback()
+        raise _http_error(exc) from exc
+
+
+@router.delete(
+    "/connectors/profiles/{profile_id}", response_model=FileConnectorProfileResponse
+)
+def deactivate_connector_profile(
+    profile_id: str,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(
+        require_any_scope(
+            "files:file:admin", "system:settings:write", "admin:settings:write"
+        )
+    ),
+):
+    try:
+        row = get_connector_profile_row(
+            session,
+            tenant_id=principal.tenant_id,
+            profile_id=profile_id,
+            include_disabled=True,
+        )
+    except FileStorageError as exc:
+        raise _http_error(exc, not_found=True) from exc
+    _require_connector_profile_write(principal, row.scope_type)
+    try:
+        changed = delete_connector_profile_row(
+            session,
+            row,
+            deletion_reason="api_delete",
+            user_id=principal.user.id,
+            api_key_id=principal.api_key.id if principal.api_key else None,
+        )
+        if changed:
+            _record_connector_settings_change(
+                session,
+                collection=FILES_CONNECTOR_PROFILES_COLLECTION,
+                resource_type=FILES_CONNECTOR_PROFILE_RESOURCE,
+                resource_id=row.id,
+                operation="deleted",
+                principal=principal,
+                tenant_id=row.tenant_id,
+                payload={
+                    "scope_type": row.scope_type,
+                    "scope_id": row.scope_id,
+                    "provider": row.provider,
+                },
+            )
+        session.commit()
+        session.refresh(row)
+        return FileConnectorProfileResponse(
+            **connector_profile_from_row(row).to_response()
+        )
+    except FileStorageError as exc:
+        session.rollback()
+        raise _http_error(exc) from exc
+    except Exception:
+        session.rollback()
+        raise

src/govoplan_files/backend/routes/connector_settings.py

@@ -0,0 +1,852 @@
+from __future__ import annotations
+
+import json
+from typing import Literal
+from fastapi import APIRouter, Depends, Query, status
+from sqlalchemy.orm import Session
+
+from govoplan_core.auth import ApiPrincipal, require_any_scope
+from govoplan_files.backend.change_tracking import (
+    FILES_CONNECTOR_CREDENTIALS_COLLECTION,
+    FILES_CONNECTOR_POLICIES_COLLECTION,
+    FILES_CONNECTOR_PROFILES_COLLECTION,
+)
+from govoplan_files.backend.schemas import (
+    FileConnectorCredentialCreateRequest,
+    FileConnectorCredentialResponse,
+    FileConnectorCredentialsResponse,
+    FileConnectorCredentialUpdateRequest,
+    FileConnectorDiscoveryRequest,
+    FileConnectorDiscoveryResponse,
+    FileConnectorSettingsDeltaResponse,
+    FileConnectorPolicyEvaluateRequest,
+    FileConnectorPolicyEvaluateResponse,
+    FileConnectorPolicyResponse,
+    FileConnectorPolicyUpdateRequest,
+    FileConnectorProfileCreateRequest,
+    FileConnectorProfileResponse,
+    FileConnectorProfilesResponse,
+    FileConnectorProviderResponse,
+    FileConnectorProvidersResponse,
+)
+from govoplan_core.db.session import get_session
+from govoplan_files.backend.storage.common import FileStorageError
+from govoplan_files.backend.storage.connector_credential_store import (
+    create_connector_credential_row,
+    get_connector_credential_row,
+    list_database_connector_credentials,
+    update_connector_credential_row,
+)
+from govoplan_files.backend.storage.connector_credential_deletion import (
+    delete_connector_credential_row,
+)
+from govoplan_files.backend.storage.connector_browse import (
+    ConnectorBrowseError,
+    ConnectorBrowseUnsupported,
+    browse_connector_profile,
+)
+from govoplan_files.backend.storage.connector_deployment import (
+    connector_effective_endpoint_url,
+    reject_api_controlled_deployment_references,
+)
+from govoplan_files.backend.storage.connector_profile_store import (
+    connector_profile_from_row,
+    create_connector_profile_row,
+)
+from govoplan_files.backend.storage.connector_providers import (
+    connector_provider_descriptors,
+)
+from govoplan_files.backend.storage.connector_policy import (
+    ConnectorAccessRequest,
+    ConnectorPolicyDenied,
+    connector_policy_decision,
+    connector_policy_sources_from_payload,
+)
+from govoplan_files.backend.storage.connector_policy_store import (
+    connector_policy_response,
+    set_connector_policy,
+)
+
+
+from govoplan_files.backend.route_support import (
+    FILES_CONNECTOR_CREDENTIAL_RESOURCE,
+    FILES_CONNECTOR_POLICY_RESOURCE,
+    FILES_CONNECTOR_PROFILE_RESOURCE,
+    _audit_connector_discovery_attempt,
+    _can_read_disabled_connector_profiles,
+    _connector_credential_response,
+    _connector_policy_error,
+    _credential_row_for_profile,
+    _discovery_profile_from_payload,
+    _ensure_campaign_file_access,
+    _ensure_connector_configuration_allowed,
+    _ensure_connector_credential_configuration_allowed,
+    _ensure_connector_local_policy_allowed,
+    _file_connector_policy_resource_id,
+    _file_connector_settings_entries,
+    _http_error,
+    _record_connector_settings_change,
+    _require_connector_credential_write,
+    _require_connector_policy_read,
+    _require_connector_profile_write,
+    _same_endpoint,
+    _visible_connector_profiles,
+    _webdav_discovery_candidates,
+)
+from govoplan_files.backend.services.connector_settings_delta import (
+    _full_file_connector_settings_delta_response,
+    _incremental_file_connector_settings_delta_response,
+)
+
+router = APIRouter(prefix="/files", tags=["files"])
+
+
+@router.post(
+    "/connector-policy/evaluate", response_model=FileConnectorPolicyEvaluateResponse
+)
+def evaluate_connector_policy(
+    payload: FileConnectorPolicyEvaluateRequest,
+    principal: ApiPrincipal = Depends(
+        require_any_scope("files:file:read", "files:file:upload", "files:file:download")
+    ),
+):
+    del principal
+    sources = connector_policy_sources_from_payload(
+        [item.model_dump(mode="json") for item in payload.policy_sources]
+    )
+    request = ConnectorAccessRequest.from_provenance(
+        payload.source_provenance.model_dump(mode="json", exclude_none=True),
+        operation=payload.operation,
+    )
+    return FileConnectorPolicyEvaluateResponse(
+        decision=connector_policy_decision(request, sources).to_dict()
+    )
+
+
+@router.get(
+    "/connectors/settings/delta", response_model=FileConnectorSettingsDeltaResponse
+)
+def connector_settings_delta(
+    scope_type: str = Query(default="tenant"),
+    scope_id: str | None = Query(default=None),
+    provider: str | None = None,
+    campaign_id: str | None = None,
+    include_disabled: bool = False,
+    include_inactive: bool = False,
+    owner_type: Literal["user", "group"] | None = None,
+    owner_id: str | None = None,
+    since: str | None = None,
+    limit: int = Query(default=100, ge=1, le=500),
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(
+        require_any_scope(
+            "files:file:admin", "system:settings:read", "admin:settings:read"
+        )
+    ),
+):
+    scope_type = scope_type.strip().casefold()
+    _require_connector_policy_read(principal, scope_type)
+    try:
+        if since is None:
+            return _full_file_connector_settings_delta_response(
+                session,
+                principal,
+                scope_type=scope_type,
+                scope_id=scope_id,
+                provider=provider,
+                campaign_id=campaign_id,
+                include_disabled=include_disabled,
+                include_inactive=include_inactive,
+                owner_type=owner_type,
+                owner_id=owner_id,
+            )
+        entries, has_more = _file_connector_settings_entries(
+            session, tenant_id=principal.tenant_id, since=since, limit=limit
+        )
+        if entries is None:
+            return _full_file_connector_settings_delta_response(
+                session,
+                principal,
+                scope_type=scope_type,
+                scope_id=scope_id,
+                provider=provider,
+                campaign_id=campaign_id,
+                include_disabled=include_disabled,
+                include_inactive=include_inactive,
+                owner_type=owner_type,
+                owner_id=owner_id,
+            )
+        return _incremental_file_connector_settings_delta_response(
+            session,
+            principal,
+            entries=entries,
+            has_more=has_more,
+            scope_type=scope_type,
+            scope_id=scope_id,
+            provider=provider,
+            campaign_id=campaign_id,
+            include_disabled=include_disabled,
+            include_inactive=include_inactive,
+            owner_type=owner_type,
+            owner_id=owner_id,
+        )
+    except (FileStorageError, ValueError, json.JSONDecodeError) as exc:
+        raise _http_error(exc) from exc
+
+
+@router.get("/connectors/providers", response_model=FileConnectorProvidersResponse)
+def list_connector_providers(
+    principal: ApiPrincipal = Depends(
+        require_any_scope("files:file:read", "files:file:upload", "files:file:admin")
+    ),
+):
+    del principal
+    return FileConnectorProvidersResponse(
+        providers=[
+            FileConnectorProviderResponse(**item.to_response())
+            for item in connector_provider_descriptors()
+        ]
+    )
+
+
+@router.post("/connectors/discover", response_model=FileConnectorDiscoveryResponse)
+def discover_connector_endpoint(
+    payload: FileConnectorDiscoveryRequest,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(
+        require_any_scope(
+            "files:file:admin", "system:settings:write", "admin:settings:write"
+        )
+    ),
+):
+    try:
+        reject_api_controlled_deployment_references(
+            password_env=payload.credentials.password_env,
+            token_env=payload.credentials.token_env,
+            secret_ref=payload.credentials.secret_ref,
+            metadata=payload.metadata,
+        )
+    except ValueError as exc:
+        raise _http_error(exc) from exc
+    if payload.provider not in {"webdav", "nextcloud"}:
+        return FileConnectorDiscoveryResponse(
+            provider=payload.provider,
+            endpoint_url=None,
+            base_path=payload.base_path,
+            status="unsupported",
+            message=f"Discovery is not implemented for {payload.provider} connectors yet",
+        )
+    candidates: list[dict[str, str]] = []
+    for endpoint_url in _webdav_discovery_candidates(payload):
+        profile = _discovery_profile_from_payload(
+            payload, endpoint_url, principal=principal
+        )
+        try:
+            _ensure_connector_configuration_allowed(
+                session,
+                principal,
+                connector_id=None,
+                credential_id=None,
+                provider=profile.provider,
+                endpoint_url=endpoint_url,
+                base_path=profile.base_path,
+                scope_type="tenant",
+                scope_id=principal.tenant_id,
+                operation="discover",
+            )
+            _audit_connector_discovery_attempt(
+                session,
+                principal,
+                provider=profile.provider,
+                endpoint_url=endpoint_url,
+                base_path=profile.base_path,
+            )
+            browse_connector_profile(profile, path=payload.base_path or "")
+        except ConnectorPolicyDenied as exc:
+            raise _connector_policy_error(exc) from exc
+        except (
+            ConnectorBrowseError,
+            ConnectorBrowseUnsupported,
+            OSError,
+            ValueError,
+            json.JSONDecodeError,
+        ) as exc:
+            message = str(exc)
+            if "credentials were rejected" in message.casefold():
+                if payload.require_valid_credentials:
+                    candidates.append(
+                        {
+                            "endpoint_url": endpoint_url,
+                            "status": "credentials_rejected",
+                            "message": "The endpoint exists, but the credentials were rejected.",
+                        }
+                    )
+                    return FileConnectorDiscoveryResponse(
+                        provider=payload.provider,
+                        endpoint_url=endpoint_url,
+                        base_path=payload.base_path,
+                        status="credentials_rejected",
+                        message="The endpoint was found, but login failed with these credentials.",
+                        candidates=candidates,
+                        metadata={"discovered_by": "webdav-auth-challenge"},
+                    )
+                candidates.append(
+                    {
+                        "endpoint_url": endpoint_url,
+                        "status": "found",
+                        "message": "The endpoint exists, but credentials are required or were rejected.",
+                    }
+                )
+                return FileConnectorDiscoveryResponse(
+                    provider=payload.provider,
+                    endpoint_url=endpoint_url,
+                    base_path=payload.base_path,
+                    status="found",
+                    message="The endpoint was found. Add working credentials before saving or testing the connection.",
+                    candidates=candidates,
+                    metadata={"discovered_by": "webdav-auth-challenge"},
+                )
+            candidates.append(
+                {"endpoint_url": endpoint_url, "status": "failed", "message": message}
+            )
+            continue
+        status_value = (
+            "usable" if _same_endpoint(endpoint_url, payload.endpoint_url) else "found"
+        )
+        message = (
+            "The supplied URL is directly usable."
+            if status_value == "usable"
+            else "A usable connector endpoint was discovered."
+        )
+        candidates.append(
+            {"endpoint_url": endpoint_url, "status": status_value, "message": message}
+        )
+        return FileConnectorDiscoveryResponse(
+            provider=payload.provider,
+            endpoint_url=endpoint_url,
+            base_path=payload.base_path,
+            status=status_value,
+            message=message,
+            candidates=candidates,
+            metadata={"discovered_by": "webdav-propfind"},
+        )
+    return FileConnectorDiscoveryResponse(
+        provider=payload.provider,
+        endpoint_url=None,
+        base_path=payload.base_path,
+        status="not_found",
+        message="No usable WebDAV endpoint was found for this server URL.",
+        candidates=candidates,
+    )
+
+
+@router.get("/connectors/credentials", response_model=FileConnectorCredentialsResponse)
+def list_connector_credentials(
+    provider: str | None = None,
+    include_disabled: bool = False,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(
+        require_any_scope(
+            "files:file:admin", "system:settings:read", "admin:settings:read"
+        )
+    ),
+):
+    provider_norm = provider.strip().casefold() if provider else None
+    credentials = list_database_connector_credentials(
+        session,
+        tenant_id=principal.tenant_id,
+        include_disabled=include_disabled
+        and _can_read_disabled_connector_profiles(principal),
+    )
+    return FileConnectorCredentialsResponse(
+        credentials=[
+            FileConnectorCredentialResponse(**credential.to_response())
+            for credential in credentials
+            if provider_norm is None or credential.provider in {None, provider_norm}
+        ]
+    )
+
+
+@router.get(
+    "/connectors/policies/{scope_type}", response_model=FileConnectorPolicyResponse
+)
+def read_connector_policy(
+    scope_type: str,
+    scope_id: str | None = Query(default=None),
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(
+        require_any_scope(
+            "files:file:admin", "system:settings:read", "admin:settings:read"
+        )
+    ),
+):
+    _require_connector_policy_read(principal, scope_type)
+    try:
+        return FileConnectorPolicyResponse(
+            **connector_policy_response(
+                session,
+                tenant_id=principal.tenant_id,
+                scope_type=scope_type,
+                scope_id=scope_id,
+            )
+        )
+    except (FileStorageError, ValueError, json.JSONDecodeError) as exc:
+        raise _http_error(exc, not_found=True) from exc
+
+
+@router.put(
+    "/connectors/policies/{scope_type}", response_model=FileConnectorPolicyResponse
+)
+def write_connector_policy(
+    scope_type: str,
+    payload: FileConnectorPolicyUpdateRequest,
+    scope_id: str | None = Query(default=None),
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(
+        require_any_scope(
+            "files:file:admin", "system:settings:write", "admin:settings:write"
+        )
+    ),
+):
+    _require_connector_profile_write(principal, scope_type)
+    try:
+        set_connector_policy(
+            session,
+            tenant_id=principal.tenant_id,
+            user_id=principal.user.id,
+            scope_type=scope_type,
+            scope_id=scope_id,
+            policy=payload.policy,
+        )
+        _record_connector_settings_change(
+            session,
+            collection=FILES_CONNECTOR_POLICIES_COLLECTION,
+            resource_type=FILES_CONNECTOR_POLICY_RESOURCE,
+            resource_id=_file_connector_policy_resource_id(scope_type, scope_id),
+            operation="updated",
+            principal=principal,
+            tenant_id=None
+            if scope_type.strip().casefold() == "system"
+            else principal.tenant_id,
+            payload={"scope_type": scope_type.strip().casefold(), "scope_id": scope_id},
+        )
+        session.commit()
+        return FileConnectorPolicyResponse(
+            **connector_policy_response(
+                session,
+                tenant_id=principal.tenant_id,
+                scope_type=scope_type,
+                scope_id=scope_id,
+            )
+        )
+    except (FileStorageError, ValueError, json.JSONDecodeError) as exc:
+        session.rollback()
+        raise _http_error(exc) from exc
+
+
+@router.post(
+    "/connectors/credentials",
+    response_model=FileConnectorCredentialResponse,
+    status_code=status.HTTP_201_CREATED,
+)
+def create_connector_credential(
+    payload: FileConnectorCredentialCreateRequest,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(
+        require_any_scope(
+            "files:file:admin", "system:settings:write", "admin:settings:write"
+        )
+    ),
+):
+    _require_connector_credential_write(principal, payload.scope_type)
+    credentials = payload.credentials
+    try:
+        _ensure_connector_credential_configuration_allowed(
+            session,
+            principal,
+            credential_id=payload.id,
+            provider=payload.provider,
+            scope_type=payload.scope_type,
+            scope_id=payload.scope_id,
+            operation="configure_credentials",
+        )
+        _ensure_connector_local_policy_allowed(
+            session,
+            principal,
+            scope_type=payload.scope_type,
+            scope_id=payload.scope_id,
+            policy=payload.policy,
+        )
+        row = create_connector_credential_row(
+            session,
+            tenant_id=principal.tenant_id,
+            user_id=principal.user.id,
+            credential_id=payload.id,
+            label=payload.label,
+            provider=payload.provider,
+            scope_type=payload.scope_type,
+            scope_id=payload.scope_id,
+            enabled=payload.enabled,
+            credential_mode=payload.credential_mode,
+            username=credentials.username,
+            password=credentials.password,
+            token=credentials.token,
+            password_env=credentials.password_env,
+            token_env=credentials.token_env,
+            secret_ref=credentials.secret_ref,
+            policy=payload.policy,
+            metadata=payload.metadata,
+        )
+        _record_connector_settings_change(
+            session,
+            collection=FILES_CONNECTOR_CREDENTIALS_COLLECTION,
+            resource_type=FILES_CONNECTOR_CREDENTIAL_RESOURCE,
+            resource_id=row.id,
+            operation="created",
+            principal=principal,
+            tenant_id=row.tenant_id,
+            payload={
+                "scope_type": row.scope_type,
+                "scope_id": row.scope_id,
+                "provider": row.provider,
+            },
+        )
+        session.commit()
+        session.refresh(row)
+        return _connector_credential_response(row)
+    except ConnectorPolicyDenied as exc:
+        session.rollback()
+        raise _connector_policy_error(exc) from exc
+    except (FileStorageError, ValueError, json.JSONDecodeError) as exc:
+        session.rollback()
+        raise _http_error(exc) from exc
+
+
+@router.get(
+    "/connectors/credentials/{credential_id}",
+    response_model=FileConnectorCredentialResponse,
+)
+def get_connector_credential(
+    credential_id: str,
+    include_disabled: bool = False,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(
+        require_any_scope(
+            "files:file:admin", "system:settings:read", "admin:settings:read"
+        )
+    ),
+):
+    try:
+        row = get_connector_credential_row(
+            session,
+            tenant_id=principal.tenant_id,
+            credential_id=credential_id,
+            include_disabled=include_disabled
+            and _can_read_disabled_connector_profiles(principal),
+        )
+        return _connector_credential_response(row)
+    except FileStorageError as exc:
+        raise _http_error(exc, not_found=True) from exc
+
+
+@router.patch(
+    "/connectors/credentials/{credential_id}",
+    response_model=FileConnectorCredentialResponse,
+)
+def update_connector_credential(
+    credential_id: str,
+    payload: FileConnectorCredentialUpdateRequest,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(
+        require_any_scope(
+            "files:file:admin", "system:settings:write", "admin:settings:write"
+        )
+    ),
+):
+    try:
+        row = get_connector_credential_row(
+            session,
+            tenant_id=principal.tenant_id,
+            credential_id=credential_id,
+            include_disabled=True,
+        )
+    except FileStorageError as exc:
+        raise _http_error(exc, not_found=True) from exc
+    _require_connector_credential_write(principal, row.scope_type)
+    credentials = payload.credentials
+    try:
+        provider = payload.provider if payload.provider is not None else row.provider
+        _ensure_connector_credential_configuration_allowed(
+            session,
+            principal,
+            credential_id=row.id,
+            provider=provider,
+            scope_type=row.scope_type,
+            scope_id=row.scope_id,
+            operation="configure_credentials",
+        )
+        if payload.policy is not None:
+            _ensure_connector_local_policy_allowed(
+                session,
+                principal,
+                scope_type=row.scope_type,
+                scope_id=row.scope_id,
+                policy=payload.policy,
+            )
+        update_connector_credential_row(
+            session,
+            row,
+            user_id=principal.user.id,
+            label=payload.label,
+            provider=payload.provider,
+            enabled=payload.enabled,
+            credential_mode=payload.credential_mode,
+            username=credentials.username if credentials else None,
+            password=credentials.password if credentials else None,
+            token=credentials.token if credentials else None,
+            password_env=credentials.password_env if credentials else None,
+            token_env=credentials.token_env if credentials else None,
+            secret_ref=credentials.secret_ref if credentials else None,
+            clear_password=payload.clear_password,
+            clear_token=payload.clear_token,
+            policy=payload.policy,
+            metadata=payload.metadata,
+        )
+        _record_connector_settings_change(
+            session,
+            collection=FILES_CONNECTOR_CREDENTIALS_COLLECTION,
+            resource_type=FILES_CONNECTOR_CREDENTIAL_RESOURCE,
+            resource_id=row.id,
+            operation="updated",
+            principal=principal,
+            tenant_id=row.tenant_id,
+            payload={
+                "scope_type": row.scope_type,
+                "scope_id": row.scope_id,
+                "provider": row.provider,
+            },
+        )
+        session.commit()
+        session.refresh(row)
+        return _connector_credential_response(row)
+    except ConnectorPolicyDenied as exc:
+        session.rollback()
+        raise _connector_policy_error(exc) from exc
+    except (FileStorageError, ValueError, json.JSONDecodeError) as exc:
+        session.rollback()
+        raise _http_error(exc) from exc
+
+
+@router.delete(
+    "/connectors/credentials/{credential_id}",
+    response_model=FileConnectorCredentialResponse,
+)
+def deactivate_connector_credential(
+    credential_id: str,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(
+        require_any_scope(
+            "files:file:admin", "system:settings:write", "admin:settings:write"
+        )
+    ),
+):
+    try:
+        row = get_connector_credential_row(
+            session,
+            tenant_id=principal.tenant_id,
+            credential_id=credential_id,
+            include_disabled=True,
+        )
+    except FileStorageError as exc:
+        raise _http_error(exc, not_found=True) from exc
+    _require_connector_credential_write(principal, row.scope_type)
+    try:
+        deletion = delete_connector_credential_row(
+            session,
+            row,
+            deletion_reason="api_delete",
+            user_id=principal.user.id,
+            api_key_id=principal.api_key.id if principal.api_key else None,
+        )
+        if deletion.changed:
+            _record_connector_settings_change(
+                session,
+                collection=FILES_CONNECTOR_CREDENTIALS_COLLECTION,
+                resource_type=FILES_CONNECTOR_CREDENTIAL_RESOURCE,
+                resource_id=row.id,
+                operation="deleted",
+                principal=principal,
+                tenant_id=row.tenant_id,
+                payload={
+                    "scope_type": row.scope_type,
+                    "scope_id": row.scope_id,
+                    "provider": row.provider,
+                },
+            )
+            for profile in deletion.affected_profiles:
+                _record_connector_settings_change(
+                    session,
+                    collection=FILES_CONNECTOR_PROFILES_COLLECTION,
+                    resource_type=FILES_CONNECTOR_PROFILE_RESOURCE,
+                    resource_id=profile.id,
+                    operation="updated",
+                    principal=principal,
+                    tenant_id=profile.tenant_id,
+                    payload={
+                        "scope_type": profile.scope_type,
+                        "scope_id": profile.scope_id,
+                        "provider": profile.provider,
+                        "reason": "credential_deleted",
+                    },
+                )
+        session.commit()
+        session.refresh(row)
+        return _connector_credential_response(row)
+    except FileStorageError as exc:
+        session.rollback()
+        raise _http_error(exc) from exc
+    except Exception:
+        session.rollback()
+        raise
+
+
+@router.get("/connectors/profiles", response_model=FileConnectorProfilesResponse)
+def list_connector_profiles(
+    provider: str | None = None,
+    campaign_id: str | None = None,
+    include_disabled: bool = False,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(
+        require_any_scope(
+            "files:file:read",
+            "files:file:upload",
+            "files:file:download",
+            "files:file:admin",
+            "system:settings:read",
+            "admin:settings:read",
+        )
+    ),
+):
+    try:
+        if campaign_id:
+            _ensure_campaign_file_access(session, principal, campaign_id)
+        profiles = _visible_connector_profiles(
+            session,
+            principal,
+            provider=provider,
+            campaign_id=campaign_id,
+            include_disabled=include_disabled
+            and _can_read_disabled_connector_profiles(principal),
+            include_admin_scopes=_can_read_disabled_connector_profiles(principal),
+            include_effective_policy=False,
+        )
+    except (OSError, ValueError, json.JSONDecodeError) as exc:
+        raise _http_error(exc) from exc
+    return FileConnectorProfilesResponse(
+        profiles=[
+            FileConnectorProfileResponse(**profile.to_response())
+            for profile in profiles
+        ]
+    )
+
+
+@router.post(
+    "/connectors/profiles",
+    response_model=FileConnectorProfileResponse,
+    status_code=status.HTTP_201_CREATED,
+)
+def create_connector_profile(
+    payload: FileConnectorProfileCreateRequest,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(
+        require_any_scope(
+            "files:file:admin", "system:settings:write", "admin:settings:write"
+        )
+    ),
+):
+    _require_connector_profile_write(principal, payload.scope_type)
+    credentials = payload.credentials
+    try:
+        credential_row = _credential_row_for_profile(
+            session,
+            principal,
+            credential_profile_id=payload.credential_profile_id,
+            provider=payload.provider,
+            profile_id=payload.id,
+            scope_type=payload.scope_type,
+            scope_id=payload.scope_id,
+        )
+        _ensure_connector_configuration_allowed(
+            session,
+            principal,
+            connector_id=payload.id,
+            credential_id=payload.credential_profile_id,
+            provider=payload.provider,
+            endpoint_url=connector_effective_endpoint_url(
+                provider=payload.provider,
+                endpoint_url=payload.endpoint_url,
+                metadata=payload.metadata,
+            ),
+            base_path=payload.base_path,
+            scope_type=payload.scope_type,
+            scope_id=payload.scope_id,
+            operation="configure",
+        )
+        _ensure_connector_local_policy_allowed(
+            session,
+            principal,
+            scope_type=payload.scope_type,
+            scope_id=payload.scope_id,
+            policy=payload.policy,
+        )
+        row = create_connector_profile_row(
+            session,
+            tenant_id=principal.tenant_id,
+            user_id=principal.user.id,
+            profile_id=payload.id,
+            label=payload.label,
+            provider=payload.provider,
+            scope_type=payload.scope_type,
+            scope_id=payload.scope_id,
+            endpoint_url=payload.endpoint_url,
+            base_path=payload.base_path,
+            enabled=payload.enabled,
+            credential_profile_id=payload.credential_profile_id,
+            credential_mode=payload.credential_mode,
+            username=credentials.username,
+            password=credentials.password,
+            token=credentials.token,
+            password_env=credentials.password_env,
+            token_env=credentials.token_env,
+            secret_ref=credentials.secret_ref,
+            capabilities=payload.capabilities,
+            policy=payload.policy,
+            metadata=payload.metadata,
+        )
+        _record_connector_settings_change(
+            session,
+            collection=FILES_CONNECTOR_PROFILES_COLLECTION,
+            resource_type=FILES_CONNECTOR_PROFILE_RESOURCE,
+            resource_id=row.id,
+            operation="created",
+            principal=principal,
+            tenant_id=row.tenant_id,
+            payload={
+                "scope_type": row.scope_type,
+                "scope_id": row.scope_id,
+                "provider": row.provider,
+            },
+        )
+        session.commit()
+        session.refresh(row)
+        return FileConnectorProfileResponse(
+            **connector_profile_from_row(
+                row, credential_row=credential_row
+            ).to_response()
+        )
+    except ConnectorPolicyDenied as exc:
+        session.rollback()
+        raise _connector_policy_error(exc) from exc
+    except (FileStorageError, ValueError, json.JSONDecodeError) as exc:
+        session.rollback()
+        raise _http_error(exc) from exc

src/govoplan_files/backend/routes/folders.py

@@ -0,0 +1,151 @@
+from __future__ import annotations
+
+from typing import Literal
+from fastapi import APIRouter, Depends, Query
+from sqlalchemy.orm import Session
+
+from govoplan_core.auth import ApiPrincipal, require_scope
+from govoplan_files.backend.schemas import (
+    FileFolderCreateRequest,
+    FileFolderDeleteRequest,
+    FileFolderDeleteResponse,
+    FileFolderResponse,
+    FileFoldersResponse,
+)
+from govoplan_core.db.session import get_session
+from govoplan_files.backend.storage.paths import UnsafeFilePathError
+from govoplan_files.backend.storage.common import FileStorageError
+from govoplan_files.backend.storage.folders import (
+    create_folder,
+    list_folders_for_user,
+    list_folders_for_user_window,
+    soft_delete_folder,
+)
+
+
+from govoplan_files.backend.route_support import (
+    _folder_response,
+    _http_error,
+    _is_admin,
+)
+from govoplan_files.backend.services.list_queries import (
+    FOLDERS_LIST_CURSOR_SCOPE,
+    _cursor_page_size,
+    _files_delta_watermark,
+    _folder_cursor_values,
+    _folders_list_fingerprint,
+    _next_folder_list_cursor,
+)
+
+router = APIRouter(prefix="/files", tags=["files"])
+
+
+@router.get("/folders", response_model=FileFoldersResponse)
+def list_file_folders(
+    owner_type: Literal["user", "group"],
+    owner_id: str,
+    page_size: int | None = Query(default=None, ge=1, le=1000),
+    cursor: str | None = None,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:read")),
+):
+    try:
+        watermark = _files_delta_watermark(session, principal.tenant_id)
+        effective_page_size = _cursor_page_size(
+            FOLDERS_LIST_CURSOR_SCOPE, cursor, page_size
+        )
+        if effective_page_size is None:
+            folders = list_folders_for_user(
+                session,
+                tenant_id=principal.tenant_id,
+                user_id=principal.user.id,
+                owner_type=owner_type,
+                owner_id=owner_id,
+                is_admin=_is_admin(principal),
+            )
+            return FileFoldersResponse(
+                folders=[_folder_response(folder) for folder in folders],
+                watermark=watermark,
+            )
+        fingerprint = _folders_list_fingerprint(
+            principal,
+            owner_type=owner_type,
+            owner_id=owner_id,
+            page_size=effective_page_size,
+        )
+        after_path, after_id = _folder_cursor_values(cursor, fingerprint=fingerprint)
+        folders, has_more = list_folders_for_user_window(
+            session,
+            tenant_id=principal.tenant_id,
+            user_id=principal.user.id,
+            owner_type=owner_type,
+            owner_id=owner_id,
+            is_admin=_is_admin(principal),
+            page_size=effective_page_size,
+            after_path=after_path,
+            after_id=after_id,
+        )
+        return FileFoldersResponse(
+            folders=[_folder_response(folder) for folder in folders],
+            cursor=cursor,
+            next_cursor=_next_folder_list_cursor(
+                principal,
+                folders,
+                owner_type=owner_type,
+                owner_id=owner_id,
+                page_size=effective_page_size,
+                has_more=has_more,
+            ),
+            watermark=watermark,
+        )
+    except FileStorageError as exc:
+        raise _http_error(exc) from exc
+
+
+@router.post("/folders", response_model=FileFolderResponse)
+def create_file_folder(
+    payload: FileFolderCreateRequest,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:organize")),
+):
+    try:
+        folder = create_folder(
+            session,
+            tenant_id=principal.tenant_id,
+            owner_type=payload.owner_type,
+            owner_id=payload.owner_id,
+            user_id=principal.user.id,
+            path=payload.path,
+            is_admin=_is_admin(principal),
+        )
+        session.commit()
+        return _folder_response(folder)
+    except (FileStorageError, UnsafeFilePathError, ValueError) as exc:
+        session.rollback()
+        raise _http_error(exc) from exc
+
+
+@router.post("/folders/delete", response_model=FileFolderDeleteResponse)
+def delete_file_folder(
+    payload: FileFolderDeleteRequest,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:delete")),
+):
+    try:
+        deleted_folders, deleted_files = soft_delete_folder(
+            session,
+            tenant_id=principal.tenant_id,
+            owner_type=payload.owner_type,
+            owner_id=payload.owner_id,
+            user_id=principal.user.id,
+            path=payload.path,
+            recursive=payload.recursive,
+            is_admin=_is_admin(principal),
+        )
+        session.commit()
+        return FileFolderDeleteResponse(
+            deleted_folders=deleted_folders, deleted_files=deleted_files
+        )
+    except (FileStorageError, UnsafeFilePathError, ValueError) as exc:
+        session.rollback()
+        raise _http_error(exc) from exc

src/govoplan_files/backend/routes/form_evidence.py

@@ -0,0 +1,172 @@
+from __future__ import annotations
+
+from datetime import UTC, datetime
+import hashlib
+
+from fastapi import (
+    APIRouter,
+    Depends,
+    File as FastAPIFile,
+    Header,
+    HTTPException,
+    UploadFile,
+    status,
+)
+from sqlalchemy.orm import Session
+
+from govoplan_core.core.events import (
+    EventActorRef,
+    EventObjectRef,
+    EventTenantRef,
+    PlatformEvent,
+    emit_platform_event,
+)
+from govoplan_core.core.institutional import EvidenceReference
+from govoplan_core.db.session import get_session
+from govoplan_files.backend.db.models import FileFormEvidenceGrant
+from govoplan_files.backend.route_support import _read_limited_upload
+from govoplan_files.backend.storage.common import FileStorageError
+from govoplan_files.backend.storage.files import create_file_asset
+
+
+router = APIRouter(prefix="/files", tags=["files"])
+
+
+@router.post(
+    "/form-evidence/upload",
+    response_model=dict[str, object],
+    status_code=status.HTTP_201_CREATED,
+)
+def upload_form_evidence(
+    file: UploadFile = FastAPIFile(...),
+    x_form_evidence_token: str = Header(alias="X-Form-Evidence-Token"),
+    session: Session = Depends(get_session),
+) -> dict[str, object]:
+    now = datetime.now(UTC)
+    try:
+        grant = (
+            session.query(FileFormEvidenceGrant)
+            .filter(
+                FileFormEvidenceGrant.token_sha256
+                == _token_sha256(x_form_evidence_token)
+            )
+            .with_for_update()
+            .one_or_none()
+        )
+        if grant is None or grant.status != "issued" or _aware(grant.expires_at) <= now:
+            raise _upload_unavailable()
+        content_type = (
+            str(file.content_type or "application/octet-stream")
+            .split(";", 1)[0]
+            .strip()
+            .casefold()
+        )
+        if grant.allowed_content_types and content_type not in set(
+            grant.allowed_content_types
+        ):
+            raise HTTPException(
+                status_code=status.HTTP_415_UNSUPPORTED_MEDIA_TYPE,
+                detail="This Form does not accept the uploaded document type.",
+            )
+        data = _read_limited_upload(file, max_bytes=grant.max_size_bytes)
+        stored = create_file_asset(
+            session,
+            tenant_id=grant.tenant_id,
+            owner_type="user",
+            owner_id=grant.custodian_user_id,
+            user_id=grant.custodian_user_id,
+            filename=file.filename or "form-attachment",
+            data=data,
+            folder=f"Form submissions/{grant.form_instance_id}",
+            content_type=content_type,
+            description="Managed attachment captured through Forms Runtime.",
+            metadata={
+                "form_evidence": {
+                    "grant_id": grant.id,
+                    "form_instance_id": grant.form_instance_id,
+                    "form_definition_id": grant.form_definition_id,
+                    "form_definition_revision": grant.form_definition_revision,
+                    "purpose": grant.purpose,
+                },
+                "source_provenance": {
+                    "source_type": "form_evidence",
+                    "connector_id": "forms_runtime.evidence.files",
+                    "provider": "forms_runtime",
+                    "external_id": grant.id,
+                    "revision": grant.form_definition_revision,
+                },
+            },
+            conflict_strategy="rename",
+            is_admin=True,
+        )
+        grant.status = "uploaded"
+        grant.file_asset_id = stored.asset.id
+        grant.file_version_id = stored.version.id
+        grant.uploaded_at = now
+        session.add(grant)
+        evidence = EvidenceReference(
+            kind="document",
+            owner_module="files",
+            evidence_id=stored.asset.id,
+            tenant_id=grant.tenant_id,
+            version=stored.version.id,
+            checksum=stored.version.checksum_sha256,
+            source_ref=(f"form_submission:{grant.form_instance_id}:grant:{grant.id}"),
+            responsible_actor_ref=f"user:{grant.custodian_user_id}",
+            captured_at=now,
+        )
+        emit_platform_event(
+            session,
+            PlatformEvent(
+                type="files.form_evidence.uploaded",
+                module_id="files",
+                payload={
+                    "grant_id": grant.id,
+                    "form_instance_id": grant.form_instance_id,
+                    "file_asset_id": stored.asset.id,
+                    "file_version_id": stored.version.id,
+                    "checksum_sha256": stored.version.checksum_sha256,
+                    "size_bytes": stored.version.size_bytes,
+                },
+                occurred_at=now,
+                actor=EventActorRef(type="user", id=grant.custodian_user_id),
+                tenant=EventTenantRef(id=grant.tenant_id),
+                resource=EventObjectRef(type="file", id=stored.asset.id),
+                classification="confidential",
+            ),
+        )
+        session.commit()
+    except HTTPException:
+        session.rollback()
+        raise
+    except (FileStorageError, ValueError) as exc:
+        session.rollback()
+        raise HTTPException(
+            status_code=status.HTTP_422_UNPROCESSABLE_CONTENT,
+            detail=str(exc),
+        ) from exc
+    return {
+        "grant_id": grant.id,
+        "evidence": evidence.to_dict(include_inspection=True),
+    }
+
+
+def _token_sha256(token: str) -> str:
+    clean = str(token or "").strip()
+    if len(clean) < 32:
+        raise _upload_unavailable()
+    return hashlib.sha256(clean.encode("utf-8")).hexdigest()
+
+
+def _upload_unavailable() -> HTTPException:
+    return HTTPException(
+        status_code=status.HTTP_404_NOT_FOUND,
+        detail="The Form evidence upload is unavailable.",
+    )
+
+
+def _aware(value: datetime) -> datetime:
+    return value if value.tzinfo is not None else value.replace(tzinfo=UTC)
+
+
+__all__ = ["router", "upload_form_evidence"]

src/govoplan_files/backend/routes/integrity.py

@@ -0,0 +1,387 @@
+from __future__ import annotations
+
+import hashlib
+
+from fastapi import APIRouter, Depends, HTTPException, Query, status
+from sqlalchemy.orm import Session
+
+from govoplan_core.auth import ApiPrincipal, require_scope
+from govoplan_core.audit.logging import audit_from_principal
+from govoplan_core.db.session import get_session
+from govoplan_files.backend.db.models import (
+    FileIntegrityFinding,
+    FileIntegrityScan,
+)
+from govoplan_files.backend.schemas import (
+    FileIntegrityActionRequest,
+    FileIntegrityActionResponse,
+    FileIntegrityFindingResponse,
+    FileIntegrityFindingsResponse,
+    FileIntegrityScanCreateRequest,
+    FileIntegrityScanRunRequest,
+    FileIntegrityScanResponse,
+    FileIntegrityScansResponse,
+)
+from govoplan_files.backend.storage.backends import StorageBackendError
+from govoplan_files.backend.storage.common import FileStorageError
+from govoplan_files.backend.storage.integrity import (
+    cleanup_orphan_finding,
+    create_integrity_scan,
+    mark_integrity_scan_failed,
+    recheck_integrity_finding,
+    run_integrity_scan_batch,
+)
+
+
+router = APIRouter(prefix="/files/integrity", tags=["files-integrity"])
+
+
+@router.get("/scans", response_model=FileIntegrityScansResponse)
+def list_integrity_scans(
+    limit: int = Query(default=50, ge=1, le=200),
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:admin")),
+) -> FileIntegrityScansResponse:
+    rows = (
+        session.query(FileIntegrityScan)
+        .filter(FileIntegrityScan.tenant_id == principal.tenant_id)
+        .order_by(FileIntegrityScan.created_at.desc(), FileIntegrityScan.id.desc())
+        .limit(limit)
+        .all()
+    )
+    return FileIntegrityScansResponse(
+        scans=[_scan_response(row) for row in rows]
+    )
+
+
+@router.post(
+    "/scans",
+    response_model=FileIntegrityScanResponse,
+    status_code=status.HTTP_201_CREATED,
+)
+def create_scan(
+    payload: FileIntegrityScanCreateRequest,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:admin")),
+) -> FileIntegrityScanResponse:
+    try:
+        scan = create_integrity_scan(
+            session,
+            tenant_id=principal.tenant_id,
+            user_id=principal.user.id,
+            verify_checksums=payload.verify_checksums,
+            batch_size=payload.batch_size,
+        )
+        audit_from_principal(
+            session,
+            principal,
+            action="files.integrity.scan_created",
+            object_type="file_integrity_scan",
+            object_id=scan.id,
+            details={
+                "storage_backend": scan.storage_backend,
+                "verify_checksums": scan.verify_checksums,
+                "batch_size": scan.batch_size,
+            },
+        )
+        session.commit()
+        return _scan_response(scan)
+    except (FileStorageError, StorageBackendError) as exc:
+        session.rollback()
+        raise _integrity_http_error(exc) from exc
+
+
+@router.post("/scans/{scan_id}/run", response_model=FileIntegrityScanResponse)
+def run_scan_batch(
+    scan_id: str,
+    payload: FileIntegrityScanRunRequest,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:admin")),
+) -> FileIntegrityScanResponse:
+    scan = _scan_for_tenant(
+        session,
+        scan_id,
+        principal.tenant_id,
+        for_update=True,
+    )
+    _assert_expected_revision(scan.revision, payload.expected_revision)
+    previous_status = scan.status
+    try:
+        run_integrity_scan_batch(session, scan)
+        if scan.status == "completed" and previous_status != "completed":
+            audit_from_principal(
+                session,
+                principal,
+                action="files.integrity.scan_completed",
+                object_type="file_integrity_scan",
+                object_id=scan.id,
+                details={
+                    "verified_blob_count": scan.verified_blob_count,
+                    "quarantined_blob_count": scan.quarantined_blob_count,
+                    "orphan_object_count": scan.orphan_object_count,
+                },
+            )
+        session.commit()
+        return _scan_response(scan)
+    except (FileStorageError, StorageBackendError) as exc:
+        session.rollback()
+        scan = _scan_for_tenant(
+            session,
+            scan_id,
+            principal.tenant_id,
+            for_update=True,
+        )
+        mark_integrity_scan_failed(scan, error=exc)
+        audit_from_principal(
+            session,
+            principal,
+            action="files.integrity.scan_failed",
+            object_type="file_integrity_scan",
+            object_id=scan.id,
+            details={"error_type": type(exc).__name__},
+        )
+        session.commit()
+        raise _integrity_http_error(exc) from exc
+
+
+@router.get(
+    "/scans/{scan_id}/findings",
+    response_model=FileIntegrityFindingsResponse,
+)
+def list_integrity_findings(
+    scan_id: str,
+    state_filter: str | None = Query(default=None, alias="state"),
+    limit: int = Query(default=500, ge=1, le=1000),
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:admin")),
+) -> FileIntegrityFindingsResponse:
+    scan = _scan_for_tenant(session, scan_id, principal.tenant_id)
+    query = session.query(FileIntegrityFinding).filter(
+        FileIntegrityFinding.scan_id == scan.id,
+        FileIntegrityFinding.tenant_id == principal.tenant_id,
+    )
+    if state_filter:
+        query = query.filter(FileIntegrityFinding.state == state_filter)
+    rows = (
+        query.order_by(
+            FileIntegrityFinding.created_at.asc(),
+            FileIntegrityFinding.id.asc(),
+        )
+        .limit(limit)
+        .all()
+    )
+    return FileIntegrityFindingsResponse(
+        findings=[_finding_response(row) for row in rows]
+    )
+
+
+@router.post(
+    "/findings/{finding_id}/recheck",
+    response_model=FileIntegrityActionResponse,
+)
+def recheck_finding(
+    finding_id: str,
+    payload: FileIntegrityActionRequest,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:admin")),
+) -> FileIntegrityActionResponse:
+    finding = _finding_for_tenant(
+        session,
+        finding_id,
+        principal.tenant_id,
+        for_update=True,
+    )
+    _assert_expected_revision(finding.revision, payload.expected_revision)
+    try:
+        result = recheck_integrity_finding(
+            session,
+            finding,
+            user_id=principal.user.id,
+            dry_run=payload.dry_run,
+        )
+        _audit_integrity_action(session, principal, result)
+        session.commit()
+        return _action_response(result)
+    except (FileStorageError, StorageBackendError) as exc:
+        session.rollback()
+        raise _integrity_http_error(exc) from exc
+
+
+@router.post(
+    "/findings/{finding_id}/cleanup",
+    response_model=FileIntegrityActionResponse,
+)
+def cleanup_finding(
+    finding_id: str,
+    payload: FileIntegrityActionRequest,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:admin")),
+) -> FileIntegrityActionResponse:
+    finding = _finding_for_tenant(
+        session,
+        finding_id,
+        principal.tenant_id,
+        for_update=True,
+    )
+    _assert_expected_revision(finding.revision, payload.expected_revision)
+    try:
+        result = cleanup_orphan_finding(
+            session,
+            finding,
+            user_id=principal.user.id,
+            dry_run=payload.dry_run,
+        )
+        _audit_integrity_action(session, principal, result)
+        session.commit()
+        return _action_response(result)
+    except (FileStorageError, StorageBackendError) as exc:
+        session.rollback()
+        raise _integrity_http_error(exc) from exc
+
+
+def _scan_for_tenant(
+    session: Session,
+    scan_id: str,
+    tenant_id: str,
+    *,
+    for_update: bool = False,
+) -> FileIntegrityScan:
+    query = session.query(FileIntegrityScan).filter(FileIntegrityScan.id == scan_id)
+    if for_update:
+        query = query.populate_existing().with_for_update()
+    scan = query.one_or_none()
+    if scan is None or scan.tenant_id != tenant_id:
+        raise HTTPException(
+            status_code=status.HTTP_404_NOT_FOUND,
+            detail="Integrity scan not found",
+        )
+    return scan
+
+
+def _finding_for_tenant(
+    session: Session,
+    finding_id: str,
+    tenant_id: str,
+    *,
+    for_update: bool = False,
+) -> FileIntegrityFinding:
+    query = session.query(FileIntegrityFinding).filter(
+        FileIntegrityFinding.id == finding_id
+    )
+    if for_update:
+        query = query.populate_existing().with_for_update()
+    finding = query.one_or_none()
+    if finding is None or finding.tenant_id != tenant_id:
+        raise HTTPException(
+            status_code=status.HTTP_404_NOT_FOUND,
+            detail="Integrity finding not found",
+        )
+    return finding
+
+
+def _assert_expected_revision(current: int, expected: int) -> None:
+    if current != expected:
+        raise HTTPException(
+            status_code=status.HTTP_409_CONFLICT,
+            detail=(
+                "The integrity record changed after it was loaded; reload before "
+                "performing this action."
+            ),
+        )
+
+
+def _audit_integrity_action(session, principal, result) -> None:
+    audit_from_principal(
+        session,
+        principal,
+        action=f"files.integrity.{result.action}",
+        object_type="file_integrity_finding",
+        object_id=result.finding.id,
+        details={
+            "scan_id": result.finding.scan_id,
+            "kind": result.finding.kind,
+            "dry_run": result.dry_run,
+            "changed": result.changed,
+            "storage_key_sha256": hashlib.sha256(
+                result.finding.storage_key.encode("utf-8")
+            ).hexdigest(),
+            "inspection": result.inspection.kind
+            if result.inspection
+            else None,
+        },
+    )
+
+
+def _scan_response(scan: FileIntegrityScan) -> FileIntegrityScanResponse:
+    return FileIntegrityScanResponse(
+        id=scan.id,
+        tenant_id=scan.tenant_id,
+        storage_backend=scan.storage_backend,
+        storage_prefix=scan.storage_prefix,
+        status=scan.status,
+        revision=scan.revision,
+        phase=scan.phase,
+        verify_checksums=scan.verify_checksums,
+        batch_size=scan.batch_size,
+        scanned_blob_count=scan.scanned_blob_count,
+        verified_blob_count=scan.verified_blob_count,
+        quarantined_blob_count=scan.quarantined_blob_count,
+        scanned_object_count=scan.scanned_object_count,
+        orphan_object_count=scan.orphan_object_count,
+        created_by_user_id=scan.created_by_user_id,
+        started_at=scan.started_at.isoformat() if scan.started_at else None,
+        completed_at=scan.completed_at.isoformat()
+        if scan.completed_at
+        else None,
+        last_error=scan.last_error,
+        created_at=scan.created_at.isoformat(),
+        updated_at=scan.updated_at.isoformat(),
+    )
+
+
+def _finding_response(
+    finding: FileIntegrityFinding,
+) -> FileIntegrityFindingResponse:
+    return FileIntegrityFindingResponse(
+        id=finding.id,
+        scan_id=finding.scan_id,
+        tenant_id=finding.tenant_id,
+        kind=finding.kind,
+        state=finding.state,
+        revision=finding.revision,
+        blob_id=finding.blob_id,
+        storage_key=finding.storage_key,
+        expected_size_bytes=finding.expected_size_bytes,
+        observed_size_bytes=finding.observed_size_bytes,
+        expected_checksum_sha256=finding.expected_checksum_sha256,
+        observed_checksum_sha256=finding.observed_checksum_sha256,
+        resolved_at=finding.resolved_at.isoformat()
+        if finding.resolved_at
+        else None,
+        resolved_by_user_id=finding.resolved_by_user_id,
+        created_at=finding.created_at.isoformat(),
+        updated_at=finding.updated_at.isoformat(),
+    )
+
+
+def _action_response(result) -> FileIntegrityActionResponse:
+    return FileIntegrityActionResponse(
+        action=result.action,
+        changed=result.changed,
+        dry_run=result.dry_run,
+        finding=_finding_response(result.finding),
+        inspection_kind=result.inspection.kind if result.inspection else None,
+        inspection_valid=result.inspection.valid if result.inspection else None,
+    )
+
+
+def _integrity_http_error(exc: Exception) -> HTTPException:
+    if isinstance(exc, FileStorageError):
+        return HTTPException(
+            status_code=status.HTTP_409_CONFLICT,
+            detail=str(exc),
+        )
+    return HTTPException(
+        status_code=status.HTTP_503_SERVICE_UNAVAILABLE,
+        detail="The configured file storage backend could not complete the integrity operation",
+    )

src/govoplan_files/backend/routes/lifecycle.py

@@ -0,0 +1,297 @@
+from __future__ import annotations
+
+from fastapi import APIRouter, Depends
+from sqlalchemy.orm import Session
+
+from govoplan_core.auth import ApiPrincipal, require_scope
+from govoplan_core.audit.logging import audit_from_principal
+from govoplan_core.db.session import get_session
+from govoplan_files.backend.route_support import (
+    _asset_response,
+    _connector_space_response,
+    _http_error,
+    _is_admin,
+)
+from govoplan_files.backend.schemas import (
+    FileAssetResponse,
+    FileBlobGcRequest,
+    FileBlobGcResponse,
+    FileConnectorSpaceResponse,
+    FileFolderDeleteRequest,
+    FileFolderRestoreResponse,
+    FileLifecycleUpdateRequest,
+    FilePurgeExecuteRequest,
+    FilePurgePreviewItem,
+    FilePurgePreviewRequest,
+    FilePurgePreviewResponse,
+    FilePurgeResponse,
+    FileRestoreResponse,
+)
+from govoplan_files.backend.storage.common import FileStorageError
+from govoplan_files.backend.storage.lifecycle import (
+    execute_asset_purge,
+    garbage_collect_unreferenced_blobs,
+    get_asset_for_lifecycle,
+    preview_asset_purge,
+    restore_asset,
+    restore_connector_space,
+    restore_folder,
+    set_asset_lifecycle,
+)
+
+
+router = APIRouter(prefix="/files", tags=["files-lifecycle"])
+
+
+@router.patch("/{file_id}/lifecycle", response_model=FileAssetResponse)
+def update_file_lifecycle(
+    file_id: str,
+    payload: FileLifecycleUpdateRequest,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:retention")),
+):
+    try:
+        asset = get_asset_for_lifecycle(
+            session,
+            tenant_id=principal.tenant_id,
+            user_id=principal.user.id,
+            asset_id=file_id,
+            is_admin=_is_admin(principal),
+        )
+        set_asset_lifecycle(
+            session,
+            asset,
+            retained_until=payload.retained_until,
+            legal_hold=payload.legal_hold,
+            reason=payload.reason,
+            expected_revision=payload.expected_revision,
+        )
+        audit_from_principal(
+            session,
+            principal,
+            action="files.lifecycle.updated",
+            object_type="file",
+            object_id=asset.id,
+            details={
+                "retained_until": (
+                    asset.retained_until.isoformat() if asset.retained_until else None
+                ),
+                "legal_hold": asset.legal_hold,
+                "lifecycle_revision": asset.lifecycle_revision,
+                "reason": asset.lifecycle_reason,
+            },
+        )
+        session.commit()
+        return _asset_response(session, asset, include_shares=True)
+    except FileStorageError as exc:
+        session.rollback()
+        raise _http_error(exc, not_found=True) from exc
+
+
+@router.post("/assets/{file_id}/restore", response_model=FileRestoreResponse)
+def restore_file(
+    file_id: str,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:restore")),
+):
+    try:
+        asset = get_asset_for_lifecycle(
+            session,
+            tenant_id=principal.tenant_id,
+            user_id=principal.user.id,
+            asset_id=file_id,
+            is_admin=_is_admin(principal),
+        )
+        changed = restore_asset(session, asset)
+        if changed:
+            audit_from_principal(
+                session,
+                principal,
+                action="files.file.restored",
+                object_type="file",
+                object_id=asset.id,
+                details={"display_path": asset.display_path},
+            )
+        session.commit()
+        return FileRestoreResponse(restored_count=1 if changed else 0)
+    except FileStorageError as exc:
+        session.rollback()
+        raise _http_error(exc, not_found=True) from exc
+
+
+@router.post("/folders/restore", response_model=FileFolderRestoreResponse)
+def restore_file_folder(
+    payload: FileFolderDeleteRequest,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:restore")),
+):
+    try:
+        restored_folders, restored_files = restore_folder(
+            session,
+            tenant_id=principal.tenant_id,
+            owner_type=payload.owner_type,
+            owner_id=payload.owner_id,
+            user_id=principal.user.id,
+            path=payload.path,
+            recursive=payload.recursive,
+            is_admin=_is_admin(principal),
+        )
+        audit_from_principal(
+            session,
+            principal,
+            action="files.folder.restored",
+            object_type="file_folder",
+            object_id=payload.path,
+            details={
+                "restored_folders": restored_folders,
+                "restored_files": restored_files,
+            },
+        )
+        session.commit()
+        return FileFolderRestoreResponse(
+            restored_folders=restored_folders, restored_files=restored_files
+        )
+    except FileStorageError as exc:
+        session.rollback()
+        raise _http_error(exc, not_found=True) from exc
+
+
+@router.post(
+    "/connector-spaces/{space_id}/restore",
+    response_model=FileConnectorSpaceResponse,
+)
+def restore_file_connector_space(
+    space_id: str,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:restore")),
+):
+    try:
+        space = restore_connector_space(
+            session,
+            tenant_id=principal.tenant_id,
+            user_id=principal.user.id,
+            space_id=space_id,
+            is_admin=_is_admin(principal),
+        )
+        audit_from_principal(
+            session,
+            principal,
+            action="files.connector_space.restored",
+            object_type="file_connector_space",
+            object_id=space.id,
+            details={"label": space.label},
+        )
+        session.commit()
+        return _connector_space_response(space)
+    except FileStorageError as exc:
+        session.rollback()
+        raise _http_error(exc, not_found=True) from exc
+
+
+@router.post("/purge/preview", response_model=FilePurgePreviewResponse)
+def preview_file_purge(
+    payload: FilePurgePreviewRequest,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:purge")),
+):
+    try:
+        preview = preview_asset_purge(
+            session, tenant_id=principal.tenant_id, file_ids=payload.file_ids
+        )
+        return FilePurgePreviewResponse(
+            preview_sha256=preview.preview_sha256,
+            eligible=preview.eligible,
+            items=[
+                FilePurgePreviewItem(
+                    file_id=item.file_id,
+                    filename=item.filename,
+                    lifecycle_revision=item.lifecycle_revision,
+                    deleted_at=(item.deleted_at.isoformat() if item.deleted_at else None),
+                    retained_until=(
+                        item.retained_until.isoformat() if item.retained_until else None
+                    ),
+                    legal_hold=item.legal_hold,
+                    blockers=list(item.blockers),
+                    blob_ids=list(item.blob_ids),
+                )
+                for item in preview.items
+            ],
+        )
+    except FileStorageError as exc:
+        raise _http_error(exc) from exc
+
+
+@router.post("/purge/execute", response_model=FilePurgeResponse)
+def execute_file_purge(
+    payload: FilePurgeExecuteRequest,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:purge")),
+):
+    try:
+        result = execute_asset_purge(
+            session,
+            tenant_id=principal.tenant_id,
+            file_ids=payload.file_ids,
+            preview_sha256=payload.preview_sha256,
+            idempotency_key=payload.idempotency_key,
+            approval_reference=payload.approval_reference,
+            before_commit=lambda assets: audit_from_principal(
+                session,
+                principal,
+                action="files.files.purged",
+                object_type="file_batch",
+                object_id=payload.preview_sha256,
+                details={
+                    "file_ids": [asset.id for asset in assets],
+                    "approval_reference": payload.approval_reference,
+                    "preview_sha256": payload.preview_sha256,
+                },
+            ),
+        )
+        return FilePurgeResponse(
+            recovery_operation_id=result.recovery_operation_id,
+            status=result.status,
+            replayed=result.replayed,
+            purged_files=result.purged_files,
+            released_blobs=result.released_blobs,
+        )
+    except FileStorageError as exc:
+        session.rollback()
+        raise _http_error(exc) from exc
+
+
+@router.post("/purge/blobs", response_model=FileBlobGcResponse)
+def garbage_collect_file_blobs(
+    payload: FileBlobGcRequest,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:purge")),
+):
+    try:
+        result = garbage_collect_unreferenced_blobs(
+            session,
+            tenant_id=principal.tenant_id,
+            limit=payload.limit,
+            approval_reference=payload.approval_reference,
+        )
+        audit_from_principal(
+            session,
+            principal,
+            action="files.blobs.garbage_collected",
+            object_type="file_blob_batch",
+            object_id=principal.tenant_id,
+            details={
+                "inspected_blobs": result.inspected_blobs,
+                "deleted_blobs": result.deleted_blobs,
+                "unresolved_operation_ids": list(result.unresolved_operation_ids),
+                "approval_reference": payload.approval_reference,
+            },
+            commit=True,
+        )
+        return FileBlobGcResponse(
+            inspected_blobs=result.inspected_blobs,
+            deleted_blobs=result.deleted_blobs,
+            unresolved_operation_ids=list(result.unresolved_operation_ids),
+        )
+    except FileStorageError as exc:
+        session.rollback()
+        raise _http_error(exc) from exc

src/govoplan_files/backend/routes/listing.py

@@ -0,0 +1,194 @@
+from __future__ import annotations
+
+from typing import Literal
+from fastapi import APIRouter, Depends, HTTPException, Query, status
+from sqlalchemy.orm import Session
+
+from govoplan_core.auth import ApiPrincipal, require_scope
+from govoplan_files.backend.schemas import (
+    FileDeltaResponse,
+    FileListResponse,
+)
+from govoplan_core.db.session import get_session
+from govoplan_files.backend.storage.files import (
+    count_assets_for_user,
+    list_assets_for_user,
+    list_assets_for_user_window,
+    list_recent_assets_for_user,
+)
+
+
+from govoplan_files.backend.route_support import (
+    _asset_list_response,
+    _ensure_campaign_file_access,
+    _ensure_list_owner_access,
+    _is_admin,
+)
+from govoplan_files.backend.services.list_queries import (
+    FILES_LIST_CURSOR_SCOPE,
+    _cursor_page_size,
+    _file_cursor_values,
+    _files_delta_response,
+    _files_delta_watermark,
+    _files_list_fingerprint,
+    _full_file_delta_response,
+    _next_file_list_cursor,
+)
+
+router = APIRouter(prefix="/files", tags=["files"])
+
+
+@router.get("/delta", response_model=FileDeltaResponse)
+def files_delta(
+    owner_type: Literal["user", "group"] | None = None,
+    owner_id: str | None = None,
+    campaign_id: str | None = None,
+    path_prefix: str | None = None,
+    since: str | None = None,
+    limit: int = Query(default=500, ge=1, le=1000),
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:read")),
+):
+    _ensure_list_owner_access(session, principal, owner_type, owner_id)
+    _ensure_campaign_file_access(session, principal, campaign_id)
+    if since is None:
+        return _full_file_delta_response(
+            session,
+            principal=principal,
+            owner_type=owner_type,
+            owner_id=owner_id,
+            campaign_id=campaign_id,
+            path_prefix=path_prefix,
+        )
+    return _files_delta_response(
+        session,
+        principal=principal,
+        owner_type=owner_type,
+        owner_id=owner_id,
+        campaign_id=campaign_id,
+        path_prefix=path_prefix,
+        since=since,
+        limit=limit,
+    )
+
+
+@router.get("", response_model=FileListResponse)
+def list_files(
+    owner_type: Literal["user", "group"] | None = None,
+    owner_id: str | None = None,
+    campaign_id: str | None = None,
+    path_prefix: str | None = None,
+    campaign_usage: Literal["linked", "unlinked"] | None = None,
+    audit_relevant: bool | None = None,
+    sort: Literal["path", "recent"] = "path",
+    page_size: int | None = Query(default=None, ge=1, le=1000),
+    cursor: str | None = None,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:read")),
+):
+    _ensure_list_owner_access(session, principal, owner_type, owner_id)
+    _ensure_campaign_file_access(session, principal, campaign_id)
+    watermark = _files_delta_watermark(session, principal.tenant_id)
+    total = count_assets_for_user(
+        session,
+        tenant_id=principal.tenant_id,
+        user_id=principal.user.id,
+        owner_type=owner_type,
+        owner_id=owner_id,
+        campaign_id=campaign_id,
+        path_prefix=path_prefix,
+        campaign_usage=campaign_usage,
+        audit_relevant=audit_relevant,
+        is_admin=_is_admin(principal),
+    )
+    if sort == "recent":
+        if cursor:
+            raise HTTPException(
+                status_code=status.HTTP_422_UNPROCESSABLE_CONTENT,
+                detail="Recent file projections do not accept a path-order cursor.",
+            )
+        recent_limit = page_size or 25
+        assets = list_recent_assets_for_user(
+            session,
+            tenant_id=principal.tenant_id,
+            user_id=principal.user.id,
+            limit=recent_limit,
+            owner_type=owner_type,
+            owner_id=owner_id,
+            campaign_id=campaign_id,
+            path_prefix=path_prefix,
+            campaign_usage=campaign_usage,
+            audit_relevant=audit_relevant,
+            is_admin=_is_admin(principal),
+        )
+        return FileListResponse(
+            files=_asset_list_response(session, assets, include_shares=True),
+            total=total,
+            watermark=watermark,
+        )
+    effective_page_size = _cursor_page_size(FILES_LIST_CURSOR_SCOPE, cursor, page_size)
+    if effective_page_size is not None:
+        fingerprint = _files_list_fingerprint(
+            principal,
+            owner_type=owner_type,
+            owner_id=owner_id,
+            campaign_id=campaign_id,
+            path_prefix=path_prefix,
+            campaign_usage=campaign_usage,
+            audit_relevant=audit_relevant,
+            page_size=effective_page_size,
+        )
+        after_display_path, after_updated_at, after_id = _file_cursor_values(
+            cursor, fingerprint=fingerprint
+        )
+        assets, has_more = list_assets_for_user_window(
+            session,
+            tenant_id=principal.tenant_id,
+            user_id=principal.user.id,
+            owner_type=owner_type,
+            owner_id=owner_id,
+            campaign_id=campaign_id,
+            path_prefix=path_prefix,
+            campaign_usage=campaign_usage,
+            audit_relevant=audit_relevant,
+            is_admin=_is_admin(principal),
+            page_size=effective_page_size,
+            after_display_path=after_display_path,
+            after_updated_at=after_updated_at,
+            after_id=after_id,
+        )
+        return FileListResponse(
+            files=_asset_list_response(session, assets, include_shares=True),
+            total=total,
+            cursor=cursor,
+            next_cursor=_next_file_list_cursor(
+                principal,
+                assets,
+                owner_type=owner_type,
+                owner_id=owner_id,
+                campaign_id=campaign_id,
+                path_prefix=path_prefix,
+                campaign_usage=campaign_usage,
+                audit_relevant=audit_relevant,
+                page_size=effective_page_size,
+                has_more=has_more,
+            ),
+            watermark=watermark,
+        )
+    assets = list_assets_for_user(
+        session,
+        tenant_id=principal.tenant_id,
+        user_id=principal.user.id,
+        owner_type=owner_type,
+        owner_id=owner_id,
+        campaign_id=campaign_id,
+        path_prefix=path_prefix,
+        campaign_usage=campaign_usage,
+        audit_relevant=audit_relevant,
+        is_admin=_is_admin(principal),
+    )
+    return FileListResponse(
+        files=_asset_list_response(session, assets, include_shares=True),
+        total=total,
+        watermark=watermark,
+    )

src/govoplan_files/backend/routes/shares.py

@@ -0,0 +1,330 @@
+from __future__ import annotations
+
+from datetime import datetime, timezone
+
+from fastapi import APIRouter, Depends, HTTPException, Query, status
+from sqlalchemy.orm import Session
+
+from govoplan_core.api.v1.schemas import (
+    ReferenceOptionListResponse,
+    ReferenceOptionResponse,
+)
+from govoplan_core.auth import ApiPrincipal, require_scope
+from govoplan_core.audit.logging import audit_from_principal
+from govoplan_core.core.references import (
+    access_scope_reference_page,
+    access_scope_reference_provider_available,
+)
+from govoplan_core.db.session import get_session
+from govoplan_files.backend.runtime import get_registry
+from govoplan_files.backend.schemas import (
+    BulkFileShareRequest,
+    BulkFileShareResponse,
+    FileShareRequest,
+    FileShareResponse,
+    FileSharesResponse,
+)
+from govoplan_files.backend.storage.common import FileStorageError
+from govoplan_files.backend.storage.files import (
+    current_file_share_for_target,
+    get_asset_for_share_management,
+    list_file_shares,
+    revoke_file_share,
+    share_file,
+    share_files,
+)
+from govoplan_files.backend.route_support import (
+    _file_share_response,
+    _http_error,
+    _is_admin,
+)
+
+router = APIRouter(prefix="/files", tags=["files"])
+
+
+@router.get(
+    "/{file_id}/share-target-options",
+    response_model=ReferenceOptionListResponse,
+)
+def search_share_targets(
+    file_id: str,
+    target_type: str,
+    q: str = "",
+    selected: list[str] = Query(default=[]),
+    limit: int = Query(default=50, ge=1, le=200),
+    cursor: str | None = None,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:share")),
+) -> ReferenceOptionListResponse:
+    if target_type not in {"user", "group"}:
+        raise HTTPException(
+            status_code=status.HTTP_422_UNPROCESSABLE_CONTENT,
+            detail="Share target type must be user or group",
+        )
+    try:
+        get_asset_for_share_management(
+            session,
+            tenant_id=principal.tenant_id,
+            user_id=principal.user.id,
+            asset_id=file_id,
+            is_admin=_is_admin(principal),
+        )
+        registry = get_registry()
+        page = access_scope_reference_page(
+            registry,
+            principal,
+            scope_type=target_type,
+            reference_kind="membership" if target_type == "user" else "group",
+            query=q,
+            selected_values=selected,
+            limit=limit,
+            cursor=cursor,
+            administrative=True,
+            session=session,
+        )
+    except ValueError as exc:
+        raise HTTPException(
+            status_code=status.HTTP_422_UNPROCESSABLE_CONTENT,
+            detail=str(exc),
+        ) from exc
+    except FileStorageError as exc:
+        raise _http_error(exc) from exc
+    return ReferenceOptionListResponse(
+        options=[
+            ReferenceOptionResponse(**option.to_dict()) for option in page.options
+        ],
+        provider_available=access_scope_reference_provider_available(registry),
+        next_cursor=page.next_cursor,
+        has_more=page.has_more,
+    )
+
+
+@router.get("/{file_id}/shares", response_model=FileSharesResponse)
+def list_shares(
+    file_id: str,
+    include_inactive: bool = False,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:share")),
+):
+    try:
+        asset = get_asset_for_share_management(
+            session,
+            tenant_id=principal.tenant_id,
+            user_id=principal.user.id,
+            asset_id=file_id,
+            is_admin=_is_admin(principal),
+        )
+        return FileSharesResponse(
+            shares=[
+                _file_share_response(share)
+                for share in list_file_shares(
+                    session,
+                    tenant_id=principal.tenant_id,
+                    asset_id=asset.id,
+                    include_inactive=include_inactive,
+                )
+            ]
+        )
+    except FileStorageError as exc:
+        raise _http_error(exc) from exc
+
+
+@router.post("/{file_id}/shares", response_model=FileShareResponse)
+def create_share(
+    file_id: str,
+    payload: FileShareRequest,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:share")),
+):
+    try:
+        asset = get_asset_for_share_management(
+            session,
+            tenant_id=principal.tenant_id,
+            user_id=principal.user.id,
+            asset_id=file_id,
+            is_admin=_is_admin(principal),
+        )
+        previous = current_file_share_for_target(
+            session,
+            tenant_id=principal.tenant_id,
+            asset_id=asset.id,
+            target_type=payload.target_type,
+            target_id=payload.target_id,
+        )
+        previous_permission = previous.permission if previous else None
+        previous_expiry = previous.expires_at if previous else None
+        share = share_file(
+            session,
+            tenant_id=principal.tenant_id,
+            asset=asset,
+            target_type=payload.target_type,
+            target_id=payload.target_id,
+            permission=payload.permission,
+            user_id=principal.user.id,
+            expires_at=payload.expires_at,
+        )
+        session.flush()
+        action = _share_audit_action(
+            existed=previous is not None,
+            previous_permission=previous_permission,
+            permission=share.permission,
+            previous_expiry=previous_expiry,
+            expiry=share.expires_at,
+        )
+        if action:
+            _audit_share_change(session, principal, share, action=action)
+        session.commit()
+        return _file_share_response(share)
+    except FileStorageError as exc:
+        session.rollback()
+        raise _http_error(exc) from exc
+
+
+@router.post("/bulk-shares", response_model=BulkFileShareResponse)
+def create_bulk_shares(
+    payload: BulkFileShareRequest,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:share")),
+):
+    try:
+        file_ids = list(dict.fromkeys(payload.file_ids))
+        assets = [
+            get_asset_for_share_management(
+                session,
+                tenant_id=principal.tenant_id,
+                user_id=principal.user.id,
+                asset_id=file_id,
+                is_admin=_is_admin(principal),
+            )
+            for file_id in file_ids
+        ]
+        previous_by_asset = {
+            asset.id: current_file_share_for_target(
+                session,
+                tenant_id=principal.tenant_id,
+                asset_id=asset.id,
+                target_type=payload.target_type,
+                target_id=payload.target_id,
+            )
+            for asset in assets
+        }
+        previous_values = {
+            asset_id: (
+                share.permission if share else None,
+                share.expires_at if share else None,
+            )
+            for asset_id, share in previous_by_asset.items()
+        }
+        shares = share_files(
+            session,
+            tenant_id=principal.tenant_id,
+            assets=assets,
+            target_type=payload.target_type,
+            target_id=payload.target_id,
+            permission=payload.permission,
+            user_id=principal.user.id,
+            expires_at=payload.expires_at,
+        )
+        session.flush()
+        for share in shares:
+            previous_permission, previous_expiry = previous_values[share.file_asset_id]
+            action = _share_audit_action(
+                existed=previous_by_asset[share.file_asset_id] is not None,
+                previous_permission=previous_permission,
+                permission=share.permission,
+                previous_expiry=previous_expiry,
+                expiry=share.expires_at,
+            )
+            if action:
+                _audit_share_change(session, principal, share, action=action)
+        session.commit()
+        return BulkFileShareResponse(
+            shared_count=len(shares),
+            shares=[_file_share_response(share) for share in shares],
+        )
+    except FileStorageError as exc:
+        session.rollback()
+        raise _http_error(exc) from exc
+
+
+@router.delete("/{file_id}/shares/{share_id}", response_model=FileShareResponse)
+def revoke_share(
+    file_id: str,
+    share_id: str,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:share")),
+):
+    try:
+        asset = get_asset_for_share_management(
+            session,
+            tenant_id=principal.tenant_id,
+            user_id=principal.user.id,
+            asset_id=file_id,
+            is_admin=_is_admin(principal),
+        )
+        share, changed = revoke_file_share(
+            session,
+            tenant_id=principal.tenant_id,
+            asset_id=asset.id,
+            share_id=share_id,
+            user_id=principal.user.id,
+        )
+        if changed:
+            _audit_share_change(
+                session, principal, share, action="files.share.revoked"
+            )
+        session.commit()
+        return _file_share_response(share)
+    except FileStorageError as exc:
+        session.rollback()
+        raise _http_error(exc) from exc
+
+
+def _share_audit_action(
+    *,
+    existed: bool,
+    previous_permission: str | None,
+    permission: str,
+    previous_expiry: datetime | None,
+    expiry: datetime | None,
+) -> str | None:
+    if not existed:
+        return "files.share.granted"
+    permission_changed = previous_permission != permission
+    expiry_changed = _normalized_expiry(previous_expiry) != _normalized_expiry(expiry)
+    if permission_changed:
+        return "files.share.changed"
+    if expiry_changed:
+        return "files.share.expiry_changed"
+    return None
+
+
+def _audit_share_change(
+    session: Session,
+    principal: ApiPrincipal,
+    share,
+    *,
+    action: str,
+) -> None:
+    audit_from_principal(
+        session,
+        principal,
+        action=action,
+        object_type="file_share",
+        object_id=share.id,
+        details={
+            "file_asset_id": share.file_asset_id,
+            "target_type": share.target_type,
+            "target_id": share.target_id,
+            "permission": share.permission,
+            "expires_at": _normalized_expiry(share.expires_at),
+        },
+    )
+
+
+def _normalized_expiry(value: datetime | None) -> str | None:
+    if value is None:
+        return None
+    if value.tzinfo is None:
+        value = value.replace(tzinfo=timezone.utc)
+    return value.astimezone(timezone.utc).isoformat()

src/govoplan_files/backend/routes/spaces.py

@@ -0,0 +1,308 @@
+from __future__ import annotations
+
+import json
+from typing import Literal
+from fastapi import APIRouter, Depends, HTTPException, status
+from sqlalchemy.orm import Session
+
+from govoplan_core.auth import ApiPrincipal, require_scope
+from govoplan_files.backend.change_tracking import (
+    FILES_CONNECTOR_SPACES_COLLECTION,
+)
+from govoplan_files.backend.schemas import (
+    FileConnectorSpaceCreateRequest,
+    FileConnectorSpaceResponse,
+    FileConnectorSpacesResponse,
+    FileConnectorSpaceUpdateRequest,
+    FileSpaceResponse,
+    FileSpacesResponse,
+)
+from govoplan_core.db.session import get_session
+from govoplan_files.backend.storage.access import group_refs_for_ids, user_group_ids
+from govoplan_files.backend.storage.common import FileStorageError
+from govoplan_files.backend.storage.connector_spaces import (
+    connector_space_owner_id,
+    create_connector_space,
+    get_connector_space_for_user,
+    list_connector_spaces_for_user,
+    soft_delete_connector_space,
+    update_connector_space,
+    validate_connector_space_write_mode,
+)
+from govoplan_files.backend.storage.connector_policy import (
+    ConnectorPolicyDenied,
+)
+
+
+from govoplan_files.backend.route_support import (
+    FILES_CONNECTOR_SPACE_RESOURCE,
+    _connector_policy_error,
+    _connector_space_file_space_response,
+    _connector_space_policy_decision,
+    _connector_space_response,
+    _http_error,
+    _is_admin,
+    _record_connector_settings_change,
+    _visible_connector_profile,
+)
+
+router = APIRouter(prefix="/files", tags=["files"])
+
+
+@router.get("/spaces", response_model=FileSpacesResponse)
+def list_file_spaces(
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:read")),
+):
+    spaces = [
+        FileSpaceResponse(
+            id=f"user:{principal.user.id}",
+            label="My files",
+            owner_type="user",
+            owner_id=principal.user.id,
+            description="Files owned by your user account.",
+        )
+    ]
+    group_ids = user_group_ids(
+        session,
+        tenant_id=principal.tenant_id,
+        user_id=principal.user.id,
+        include_admin_groups=_is_admin(principal),
+    )
+    if group_ids:
+        groups = group_refs_for_ids(tenant_id=principal.tenant_id, group_ids=group_ids)
+        spaces.extend(
+            FileSpaceResponse(
+                id=f"group:{group.id}",
+                label=f"{group.name} files",
+                owner_type="group",
+                owner_id=group.id,
+                description="Files owned by this group.",
+            )
+            for group in groups
+        )
+    connector_spaces = list_connector_spaces_for_user(
+        session,
+        tenant_id=principal.tenant_id,
+        user_id=principal.user.id,
+        is_admin=_is_admin(principal),
+    )
+    spaces.extend(
+        _connector_space_file_space_response(space) for space in connector_spaces
+    )
+    return FileSpacesResponse(spaces=spaces)
+
+
+@router.get("/connector-spaces", response_model=FileConnectorSpacesResponse)
+def list_file_connector_spaces(
+    owner_type: Literal["user", "group"] | None = None,
+    owner_id: str | None = None,
+    include_inactive: bool = False,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:read")),
+):
+    try:
+        spaces = list_connector_spaces_for_user(
+            session,
+            tenant_id=principal.tenant_id,
+            user_id=principal.user.id,
+            owner_type=owner_type,
+            owner_id=owner_id,
+            include_inactive=include_inactive and _is_admin(principal),
+            is_admin=_is_admin(principal),
+        )
+        return FileConnectorSpacesResponse(
+            spaces=[_connector_space_response(space) for space in spaces]
+        )
+    except FileStorageError as exc:
+        raise _http_error(exc) from exc
+
+
+@router.post(
+    "/connector-spaces",
+    response_model=FileConnectorSpaceResponse,
+    status_code=status.HTTP_201_CREATED,
+)
+def create_file_connector_space(
+    payload: FileConnectorSpaceCreateRequest,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:organize")),
+):
+    if payload.owner_type == "group" and not payload.owner_id:
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST,
+            detail="owner_id is required for group connector spaces",
+        )
+    target_owner = payload.owner_id or principal.user.id
+    try:
+        profile = _visible_connector_profile(
+            session, principal, payload.connector_profile_id
+        )
+        decision = _connector_space_policy_decision(
+            profile,
+            library_id=payload.library_id,
+            remote_path=payload.remote_path,
+            operation="link",
+        )
+        if not decision.allowed:
+            raise ConnectorPolicyDenied(decision)
+        validate_connector_space_write_mode(profile, read_only=payload.read_only)
+        space = create_connector_space(
+            session,
+            tenant_id=principal.tenant_id,
+            owner_type=payload.owner_type,
+            owner_id=target_owner,
+            user_id=principal.user.id,
+            label=payload.label,
+            profile=profile,
+            library_id=payload.library_id,
+            remote_path=payload.remote_path,
+            sync_mode=payload.sync_mode,
+            read_only=payload.read_only,
+            metadata=payload.metadata,
+            is_admin=_is_admin(principal),
+        )
+        _record_connector_settings_change(
+            session,
+            collection=FILES_CONNECTOR_SPACES_COLLECTION,
+            resource_type=FILES_CONNECTOR_SPACE_RESOURCE,
+            resource_id=space.id,
+            operation="created",
+            principal=principal,
+            tenant_id=space.tenant_id,
+            payload={
+                "owner_type": space.owner_type,
+                "owner_id": connector_space_owner_id(space),
+            },
+        )
+        session.commit()
+        return _connector_space_response(space)
+    except ConnectorPolicyDenied as exc:
+        session.rollback()
+        raise _connector_policy_error(exc) from exc
+    except (FileStorageError, ValueError, json.JSONDecodeError) as exc:
+        session.rollback()
+        raise _http_error(exc) from exc
+
+
+@router.patch("/connector-spaces/{space_id}", response_model=FileConnectorSpaceResponse)
+def update_file_connector_space(
+    space_id: str,
+    payload: FileConnectorSpaceUpdateRequest,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:organize")),
+):
+    try:
+        space = get_connector_space_for_user(
+            session,
+            tenant_id=principal.tenant_id,
+            user_id=principal.user.id,
+            space_id=space_id,
+            include_inactive=_is_admin(principal),
+            is_admin=_is_admin(principal),
+        )
+    except FileStorageError as exc:
+        raise _http_error(exc, not_found=True) from exc
+    try:
+        if (
+            payload.library_id is not None
+            or payload.remote_path is not None
+            or payload.read_only is not None
+        ):
+            profile = _visible_connector_profile(
+                session, principal, space.connector_profile_id
+            )
+            decision = _connector_space_policy_decision(
+                profile,
+                library_id=payload.library_id
+                if payload.library_id is not None
+                else space.library_id,
+                remote_path=payload.remote_path
+                if payload.remote_path is not None
+                else space.remote_path,
+                operation="link",
+            )
+            if not decision.allowed:
+                raise ConnectorPolicyDenied(decision)
+            validate_connector_space_write_mode(
+                profile,
+                read_only=(
+                    payload.read_only
+                    if payload.read_only is not None
+                    else space.read_only
+                ),
+            )
+        update_connector_space(
+            session,
+            space,
+            user_id=principal.user.id,
+            label=payload.label,
+            library_id=payload.library_id,
+            remote_path=payload.remote_path,
+            sync_mode=payload.sync_mode,
+            read_only=payload.read_only,
+            is_active=payload.is_active,
+            metadata=payload.metadata,
+            is_admin=_is_admin(principal),
+        )
+        _record_connector_settings_change(
+            session,
+            collection=FILES_CONNECTOR_SPACES_COLLECTION,
+            resource_type=FILES_CONNECTOR_SPACE_RESOURCE,
+            resource_id=space.id,
+            operation="updated",
+            principal=principal,
+            tenant_id=space.tenant_id,
+            payload={
+                "owner_type": space.owner_type,
+                "owner_id": connector_space_owner_id(space),
+            },
+        )
+        session.commit()
+        return _connector_space_response(space)
+    except ConnectorPolicyDenied as exc:
+        session.rollback()
+        raise _connector_policy_error(exc) from exc
+    except (FileStorageError, ValueError, json.JSONDecodeError) as exc:
+        session.rollback()
+        raise _http_error(exc) from exc
+
+
+@router.delete(
+    "/connector-spaces/{space_id}", response_model=FileConnectorSpaceResponse
+)
+def delete_file_connector_space(
+    space_id: str,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:organize")),
+):
+    try:
+        space = get_connector_space_for_user(
+            session,
+            tenant_id=principal.tenant_id,
+            user_id=principal.user.id,
+            space_id=space_id,
+            include_inactive=True,
+            is_admin=_is_admin(principal),
+        )
+        soft_delete_connector_space(
+            session, space, user_id=principal.user.id, is_admin=_is_admin(principal)
+        )
+        _record_connector_settings_change(
+            session,
+            collection=FILES_CONNECTOR_SPACES_COLLECTION,
+            resource_type=FILES_CONNECTOR_SPACE_RESOURCE,
+            resource_id=space.id,
+            operation="deleted",
+            principal=principal,
+            tenant_id=space.tenant_id,
+            payload={
+                "owner_type": space.owner_type,
+                "owner_id": connector_space_owner_id(space),
+            },
+        )
+        session.commit()
+        return _connector_space_response(space)
+    except FileStorageError as exc:
+        session.rollback()
+        raise _http_error(exc, not_found=True) from exc

src/govoplan_files/backend/routes/transfers.py

@@ -0,0 +1,213 @@
+from __future__ import annotations
+
+import tempfile
+from fastapi import APIRouter, Depends
+from fastapi.responses import FileResponse
+from starlette.background import BackgroundTask
+from sqlalchemy.orm import Session
+
+from govoplan_core.auth import ApiPrincipal, require_scope
+from govoplan_files.backend.schemas import (
+    ArchiveRequest,
+    PatternMatchResponse,
+    PatternResolveRequest,
+    PatternResolveResponse,
+    RenamePreviewItem,
+    RenameRequest,
+    RenameResponse,
+    TransferRequest,
+    TransferResponse,
+    _conflict_resolutions,
+)
+from govoplan_core.db.session import get_session
+from govoplan_files.backend.storage.paths import (
+    UnsafeFilePathError,
+    filename_from_path,
+    normalize_logical_path,
+)
+from govoplan_files.backend.storage.archives import create_zip_file
+from govoplan_files.backend.storage.common import FileStorageError
+from govoplan_files.backend.storage.files import (
+    get_asset_for_user,
+    list_assets_for_user,
+)
+from govoplan_files.backend.storage.search import resolve_patterns
+from govoplan_files.backend.storage.transfers import (
+    rename_selection,
+    transfer_selection,
+)
+
+
+from govoplan_files.backend.route_support import (
+    _asset_response,
+    _attachment_disposition,
+    _audit_connector_access,
+    _cleanup_temp_file,
+    _ensure_campaign_file_access,
+    _ensure_list_owner_access,
+    _http_error,
+    _is_admin,
+)
+
+router = APIRouter(prefix="/files", tags=["files"])
+
+
+@router.post("/bulk-rename", response_model=RenameResponse)
+def bulk_rename(
+    payload: RenameRequest,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:organize")),
+):
+    try:
+        plan = rename_selection(
+            session,
+            tenant_id=principal.tenant_id,
+            user_id=principal.user.id,
+            file_ids=payload.file_ids,
+            folder_paths=payload.folder_paths,
+            owner_type=payload.owner_type,
+            owner_id=payload.owner_id,
+            mode=payload.mode,
+            new_name=payload.new_name,
+            find=payload.find,
+            replacement=payload.replacement,
+            prefix=payload.prefix,
+            suffix=payload.suffix,
+            recursive=payload.recursive,
+            dry_run=payload.dry_run,
+            is_admin=_is_admin(principal),
+        )
+        if not payload.dry_run:
+            session.commit()
+        return RenameResponse(
+            dry_run=payload.dry_run,
+            items=[
+                RenamePreviewItem(
+                    kind=item.kind,
+                    id=item.id,
+                    file_id=item.id if item.kind == "file" else None,
+                    folder_path=item.old_path if item.kind == "folder" else None,
+                    old_path=item.old_path,
+                    new_path=item.new_path,
+                )
+                for item in plan
+            ],
+        )
+    except (FileStorageError, UnsafeFilePathError, ValueError) as exc:
+        session.rollback()
+        raise _http_error(exc) from exc
+
+
+@router.post("/transfer", response_model=TransferResponse)
+def transfer_files(
+    payload: TransferRequest,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:organize")),
+):
+    try:
+        files, folders = transfer_selection(
+            session,
+            tenant_id=principal.tenant_id,
+            user_id=principal.user.id,
+            operation=payload.operation,
+            file_ids=payload.file_ids,
+            folder_paths=payload.folder_paths,
+            source_owner_type=payload.source_owner_type,
+            source_owner_id=payload.source_owner_id,
+            target_owner_type=payload.target_owner_type,
+            target_owner_id=payload.target_owner_id,
+            target_folder=payload.target_folder,
+            conflict_strategy=payload.conflict_strategy,
+            conflict_resolutions=_conflict_resolutions(payload.conflict_resolutions),
+            is_admin=_is_admin(principal),
+        )
+        session.commit()
+        return TransferResponse(
+            operation=payload.operation, files=files, folders=folders
+        )
+    except (FileStorageError, UnsafeFilePathError, ValueError) as exc:
+        session.rollback()
+        raise _http_error(exc) from exc
+
+
+@router.post("/archive.zip")
+def download_archive(
+    payload: ArchiveRequest,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:download")),
+):
+    try:
+        assets = [
+            get_asset_for_user(
+                session,
+                tenant_id=principal.tenant_id,
+                user_id=principal.user.id,
+                asset_id=file_id,
+                is_admin=_is_admin(principal),
+            )
+            for file_id in payload.file_ids
+        ]
+        tmp = tempfile.NamedTemporaryFile(
+            prefix="govoplan-files-", suffix=".zip", delete=False
+        )
+        tmp_path = tmp.name
+        tmp.close()
+        try:
+            create_zip_file(session, assets, tmp_path)
+        except Exception:
+            _cleanup_temp_file(tmp_path)
+            raise
+        _audit_connector_access(session, principal, assets, operation="archive")
+    except FileStorageError as exc:
+        raise _http_error(exc) from exc
+    filename = filename_from_path(
+        normalize_logical_path(payload.filename, fallback_filename="files.zip")
+    )
+    headers = {"Content-Disposition": _attachment_disposition(filename)}
+    return FileResponse(
+        tmp_path,
+        media_type="application/zip",
+        headers=headers,
+        background=BackgroundTask(_cleanup_temp_file, tmp_path),
+    )
+
+
+@router.post("/resolve-patterns", response_model=PatternResolveResponse)
+def resolve_file_patterns(
+    payload: PatternResolveRequest,
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:read")),
+):
+    _ensure_list_owner_access(session, principal, payload.owner_type, payload.owner_id)
+    _ensure_campaign_file_access(session, principal, payload.campaign_id)
+    try:
+        assets = list_assets_for_user(
+            session,
+            tenant_id=principal.tenant_id,
+            user_id=principal.user.id,
+            owner_type=payload.owner_type,
+            owner_id=payload.owner_id,
+            campaign_id=payload.campaign_id,
+            path_prefix=payload.path_prefix,
+            is_admin=_is_admin(principal),
+        )
+        resolved, unmatched = resolve_patterns(
+            assets,
+            payload.patterns,
+            base_path=payload.path_prefix,
+            case_sensitive=payload.case_sensitive,
+        )
+        return PatternResolveResponse(
+            patterns=[
+                PatternMatchResponse(
+                    pattern=item.pattern,
+                    matches=[_asset_response(session, asset) for asset in item.matches],
+                )
+                for item in resolved
+            ],
+            unmatched=[_asset_response(session, asset) for asset in unmatched]
+            if payload.include_unmatched
+            else [],
+        )
+    except (FileStorageError, UnsafeFilePathError, ValueError) as exc:
+        raise _http_error(exc) from exc

src/govoplan_files/backend/routes/uploads.py

@@ -0,0 +1,487 @@
+from __future__ import annotations
+
+import hashlib
+import hmac
+import json
+from datetime import UTC, datetime, timedelta
+from typing import Literal
+from fastapi import APIRouter, Depends, File as FastAPIFile, Form, UploadFile
+from sqlalchemy.orm import Session
+
+from govoplan_core.auth import ApiPrincipal, require_scope
+from govoplan_core.security.secrets import (
+    TransientPayloadError,
+    open_transient_payload,
+    seal_transient_payload,
+)
+from govoplan_files.backend.schemas import (
+    ArchiveEntryResponse,
+    ArchivePreviewResponse,
+    ConflictResolutionRequest,
+    FileUploadResponse,
+    _conflict_resolutions,
+)
+from govoplan_files.backend.db.models import FileAsset
+from govoplan_core.db.session import get_session
+from govoplan_files.backend.runtime import settings
+from govoplan_files.backend.storage.paths import (
+    UnsafeFilePathError,
+    normalize_folder,
+)
+from govoplan_files.backend.storage.archives import (
+    archive_format_for_filename,
+    extract_archive_upload,
+    extract_zip_upload,
+    inspect_archive,
+)
+from govoplan_files.backend.storage.common import FileStorageError
+from govoplan_files.backend.storage.connector_policy import (
+    ConnectorPolicyDenied,
+)
+from govoplan_files.backend.storage.files import (
+    create_file_asset,
+)
+
+
+from govoplan_files.backend.route_support import (
+    _asset_response,
+    _audit_connector_imports,
+    _cleanup_temp_file,
+    _connector_policy_error,
+    _enforce_connector_policy,
+    _http_error,
+    _is_admin,
+    _read_limited_upload,
+    _source_metadata_from_form,
+    _spool_limited_upload_to_temp,
+)
+
+router = APIRouter(prefix="/files", tags=["files"])
+_ARCHIVE_PREVIEW_PURPOSE = "files.archive-preview.v1"
+
+
+def _archive_suffix(filename: str) -> str:
+    lowered = filename.casefold()
+    for suffix in (".tar.bz2", ".tar.gz", ".tar.xz", ".tbz2", ".tgz", ".txz", ".tar", ".zip"):
+        if lowered.endswith(suffix):
+            return suffix
+    return ".archive"
+
+
+def _archive_sha256(path: str) -> str:
+    digest = hashlib.sha256()
+    with open(path, "rb") as source:
+        while chunk := source.read(1024 * 1024):
+            digest.update(chunk)
+    return digest.hexdigest()
+
+
+def _archive_selected_paths(value: str) -> list[str]:
+    parsed = json.loads(value)
+    if not isinstance(parsed, list) or not parsed:
+        raise FileStorageError("Select at least one archive file or folder")
+    if len(parsed) > settings.file_archive_max_entries:
+        raise FileStorageError(
+            "Archive selection exceeds the configured entry limit"
+        )
+    selected: list[str] = []
+    for item in parsed:
+        if not isinstance(item, str) or not item.strip():
+            raise FileStorageError("Archive selection contains an invalid path")
+        if len(item) > 4096:
+            raise FileStorageError("Archive selection path is too long")
+        selected.append(item)
+    return selected
+
+
+def _validate_archive_preview_token(
+    token: str,
+    *,
+    tenant_id: str,
+    user_id: str,
+    owner_type: str,
+    owner_id: str,
+    path: str,
+    campaign_id: str | None,
+    archive_format: str,
+    archive_sha256: str,
+) -> None:
+    payload = open_transient_payload(
+        token,
+        ttl_seconds=settings.file_archive_preview_ttl_seconds,
+    )
+    expected = {
+        "purpose": _ARCHIVE_PREVIEW_PURPOSE,
+        "tenant_id": tenant_id,
+        "user_id": user_id,
+        "owner_type": owner_type,
+        "owner_id": owner_id,
+        "path": path,
+        "campaign_id": campaign_id or "",
+        "archive_format": archive_format,
+    }
+    if any(payload.get(key) != value for key, value in expected.items()):
+        raise FileStorageError(
+            "Archive preview does not match this upload destination"
+        )
+    token_digest = payload.get("archive_sha256")
+    if not isinstance(token_digest, str) or not hmac.compare_digest(
+        token_digest, archive_sha256
+    ):
+        raise FileStorageError(
+            "Archive contents changed after preview; preview it again"
+        )
+
+
+@router.post("/archive-preview", response_model=ArchivePreviewResponse)
+def preview_archive_upload(
+    file: UploadFile = FastAPIFile(...),
+    owner_type: Literal["user", "group"] = Form(default="user"),
+    owner_id: str | None = Form(default=None),
+    path: str = Form(default=""),
+    campaign_id: str | None = Form(default=None),
+    password: str | None = Form(default=None),
+    principal: ApiPrincipal = Depends(require_scope("files:file:upload")),
+):
+    target_owner = owner_id or principal.user.id
+    archive_path: str | None = None
+    filename = file.filename or "archive"
+    try:
+        archive_format = archive_format_for_filename(filename)
+        archive_path = _spool_limited_upload_to_temp(
+            file,
+            max_bytes=settings.file_upload_zip_max_bytes,
+            suffix=_archive_suffix(filename),
+        )
+        inspection = inspect_archive(
+            archive_path,
+            filename=filename,
+            password=password,
+            max_entries=settings.file_archive_max_entries,
+            max_expanded_bytes=settings.file_archive_max_expanded_bytes,
+            max_expansion_ratio=settings.file_archive_max_expansion_ratio,
+        )
+        digest = _archive_sha256(archive_path)
+        normalized_path = normalize_folder(path)
+        preview_token = seal_transient_payload(
+            {
+                "purpose": _ARCHIVE_PREVIEW_PURPOSE,
+                "tenant_id": principal.tenant_id,
+                "user_id": principal.user.id,
+                "owner_type": owner_type,
+                "owner_id": target_owner,
+                "path": normalized_path,
+                "campaign_id": campaign_id or "",
+                "archive_format": archive_format,
+                "archive_sha256": digest,
+            }
+        )
+        expires_at = datetime.now(UTC) + timedelta(
+            seconds=settings.file_archive_preview_ttl_seconds
+        )
+        return ArchivePreviewResponse(
+            preview_token=preview_token,
+            archive_format=inspection.archive_format,
+            entries=[
+                ArchiveEntryResponse(
+                    path=entry.path,
+                    kind=entry.kind,
+                    size_bytes=entry.size_bytes,
+                    compressed_size_bytes=entry.compressed_size_bytes,
+                    encrypted=entry.encrypted,
+                )
+                for entry in inspection.entries
+            ],
+            file_count=inspection.file_count,
+            directory_count=inspection.directory_count,
+            expanded_size_bytes=inspection.expanded_size_bytes,
+            compressed_size_bytes=inspection.compressed_size_bytes,
+            requires_password=inspection.requires_password,
+            password_verified=inspection.password_verified,
+            expires_at=expires_at.isoformat(),
+        )
+    except (FileStorageError, UnsafeFilePathError, ValueError) as exc:
+        raise _http_error(exc) from exc
+    finally:
+        if archive_path:
+            _cleanup_temp_file(archive_path)
+
+
+@router.post("/archive-confirm", response_model=FileUploadResponse)
+def confirm_archive_upload(
+    file: UploadFile = FastAPIFile(...),
+    preview_token: str = Form(...),
+    selected_paths_json: str = Form(...),
+    owner_type: Literal["user", "group"] = Form(default="user"),
+    owner_id: str | None = Form(default=None),
+    path: str = Form(default=""),
+    campaign_id: str | None = Form(default=None),
+    password: str | None = Form(default=None),
+    conflict_strategy: Literal["reject", "overwrite", "rename"] = Form(
+        default="reject"
+    ),
+    conflict_resolutions_json: str | None = Form(default=None),
+    source_provenance_json: str | None = Form(default=None),
+    source_revision: str | None = Form(default=None),
+    connector_policy_json: str | None = Form(default=None),
+    encryption_vault_id: str | None = Form(default=None),
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:upload")),
+):
+    target_owner = owner_id or principal.user.id
+    archive_path: str | None = None
+    filename = file.filename or "archive"
+    try:
+        raw_resolutions = (
+            json.loads(conflict_resolutions_json)
+            if conflict_resolutions_json
+            else []
+        )
+        upload_resolutions = _conflict_resolutions(
+            [ConflictResolutionRequest(**item) for item in raw_resolutions]
+        )
+        selected_paths = _archive_selected_paths(selected_paths_json)
+        _enforce_connector_policy(
+            source_provenance_json,
+            connector_policy_json,
+            operation="import",
+        )
+        metadata = _source_metadata_from_form(
+            source_provenance_json, source_revision
+        )
+        archive_format = archive_format_for_filename(filename)
+        normalized_path = normalize_folder(path)
+        archive_path = _spool_limited_upload_to_temp(
+            file,
+            max_bytes=settings.file_upload_zip_max_bytes,
+            suffix=_archive_suffix(filename),
+        )
+        digest = _archive_sha256(archive_path)
+        _validate_archive_preview_token(
+            preview_token,
+            tenant_id=principal.tenant_id,
+            user_id=principal.user.id,
+            owner_type=owner_type,
+            owner_id=target_owner,
+            path=normalized_path,
+            campaign_id=campaign_id,
+            archive_format=archive_format,
+            archive_sha256=digest,
+        )
+        extracted = extract_archive_upload(
+            session,
+            tenant_id=principal.tenant_id,
+            owner_type=owner_type,
+            owner_id=target_owner,
+            user_id=principal.user.id,
+            archive_data=archive_path,
+            filename=filename,
+            folder=normalized_path,
+            campaign_id=campaign_id,
+            selected_paths=selected_paths,
+            password=password,
+            conflict_strategy=conflict_strategy,
+            conflict_resolutions=upload_resolutions,
+            metadata=metadata,
+            is_admin=_is_admin(principal),
+            encryption_vault_id=encryption_vault_id,
+            max_entries=settings.file_archive_max_entries,
+            max_file_bytes=settings.file_upload_max_bytes,
+            max_expanded_bytes=settings.file_archive_max_expanded_bytes,
+            max_expansion_ratio=settings.file_archive_max_expansion_ratio,
+        )
+        uploaded_assets = [item.asset for item in extracted]
+        _audit_connector_imports(session, principal, uploaded_assets)
+        session.commit()
+        return FileUploadResponse(
+            files=[
+                _asset_response(session, asset, include_shares=True)
+                for asset in uploaded_assets
+            ]
+        )
+    except ConnectorPolicyDenied as exc:
+        session.rollback()
+        raise _connector_policy_error(exc) from exc
+    except (
+        FileStorageError,
+        TransientPayloadError,
+        UnsafeFilePathError,
+        ValueError,
+        json.JSONDecodeError,
+    ) as exc:
+        session.rollback()
+        raise _http_error(exc) from exc
+    finally:
+        if archive_path:
+            _cleanup_temp_file(archive_path)
+
+
+@router.post("/upload", response_model=FileUploadResponse)
+def upload_files(
+    files: list[UploadFile] = FastAPIFile(...),
+    owner_type: Literal["user", "group"] = Form(default="user"),
+    owner_id: str | None = Form(default=None),
+    path: str = Form(default=""),
+    campaign_id: str | None = Form(default=None),
+    unpack_zip: bool = Form(default=False),
+    conflict_strategy: Literal["reject", "overwrite", "rename"] = Form(
+        default="reject"
+    ),
+    conflict_resolutions_json: str | None = Form(default=None),
+    source_provenance_json: str | None = Form(default=None),
+    source_revision: str | None = Form(default=None),
+    connector_policy_json: str | None = Form(default=None),
+    encryption_vault_id: str | None = Form(default=None),
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:upload")),
+):
+    target_owner = owner_id or principal.user.id
+    uploaded_assets: list[FileAsset] = []
+    try:
+        raw_resolutions = (
+            json.loads(conflict_resolutions_json) if conflict_resolutions_json else []
+        )
+        upload_resolutions = _conflict_resolutions(
+            [ConflictResolutionRequest(**item) for item in raw_resolutions]
+        )
+        _enforce_connector_policy(
+            source_provenance_json, connector_policy_json, operation="import"
+        )
+        metadata = _source_metadata_from_form(source_provenance_json, source_revision)
+        for upload in files:
+            filename = upload.filename or "file"
+            content_type = upload.content_type or None
+            upload_limit = (
+                settings.file_upload_zip_max_bytes
+                if unpack_zip and filename.lower().endswith(".zip")
+                else settings.file_upload_max_bytes
+            )
+            if unpack_zip and filename.lower().endswith(".zip"):
+                zip_path = _spool_limited_upload_to_temp(
+                    upload, max_bytes=upload_limit, suffix=".zip"
+                )
+                try:
+                    extracted = extract_zip_upload(
+                        session,
+                        tenant_id=principal.tenant_id,
+                        owner_type=owner_type,
+                        owner_id=target_owner,
+                        user_id=principal.user.id,
+                        zip_data=zip_path,
+                        folder=path,
+                        campaign_id=campaign_id,
+                        conflict_strategy=conflict_strategy,
+                        conflict_resolutions=upload_resolutions,
+                        metadata=metadata,
+                        is_admin=_is_admin(principal),
+                        encryption_vault_id=encryption_vault_id,
+                        max_file_bytes=settings.file_upload_max_bytes,
+                        max_total_bytes=settings.file_upload_zip_max_bytes,
+                    )
+                finally:
+                    _cleanup_temp_file(zip_path)
+                uploaded_assets.extend(item.asset for item in extracted)
+                continue
+            data = _read_limited_upload(upload, max_bytes=upload_limit)
+            stored = create_file_asset(
+                session,
+                tenant_id=principal.tenant_id,
+                owner_type=owner_type,
+                owner_id=target_owner,
+                user_id=principal.user.id,
+                filename=filename,
+                data=data,
+                folder=path,
+                content_type=content_type,
+                campaign_id=campaign_id,
+                conflict_strategy=conflict_strategy,
+                conflict_resolutions=upload_resolutions,
+                metadata=metadata,
+                is_admin=_is_admin(principal),
+                encryption_vault_id=encryption_vault_id,
+            )
+            uploaded_assets.append(stored.asset)
+        _audit_connector_imports(session, principal, uploaded_assets)
+        session.commit()
+    except ConnectorPolicyDenied as exc:
+        session.rollback()
+        raise _connector_policy_error(exc) from exc
+    except (FileStorageError, UnsafeFilePathError, ValueError) as exc:
+        session.rollback()
+        raise _http_error(exc) from exc
+    return FileUploadResponse(
+        files=[
+            _asset_response(session, asset, include_shares=True)
+            for asset in uploaded_assets
+        ]
+    )
+
+
+@router.post("/upload-zip", response_model=FileUploadResponse)
+def upload_zip(
+    file: UploadFile = FastAPIFile(...),
+    owner_type: Literal["user", "group"] = Form(default="user"),
+    owner_id: str | None = Form(default=None),
+    path: str = Form(default=""),
+    campaign_id: str | None = Form(default=None),
+    conflict_strategy: Literal["reject", "overwrite", "rename"] = Form(
+        default="reject"
+    ),
+    conflict_resolutions_json: str | None = Form(default=None),
+    source_provenance_json: str | None = Form(default=None),
+    source_revision: str | None = Form(default=None),
+    connector_policy_json: str | None = Form(default=None),
+    encryption_vault_id: str | None = Form(default=None),
+    session: Session = Depends(get_session),
+    principal: ApiPrincipal = Depends(require_scope("files:file:upload")),
+):
+    target_owner = owner_id or principal.user.id
+    zip_path: str | None = None
+    try:
+        raw_resolutions = (
+            json.loads(conflict_resolutions_json) if conflict_resolutions_json else []
+        )
+        upload_resolutions = _conflict_resolutions(
+            [ConflictResolutionRequest(**item) for item in raw_resolutions]
+        )
+        _enforce_connector_policy(
+            source_provenance_json, connector_policy_json, operation="import"
+        )
+        metadata = _source_metadata_from_form(source_provenance_json, source_revision)
+        zip_path = _spool_limited_upload_to_temp(
+            file, max_bytes=settings.file_upload_zip_max_bytes, suffix=".zip"
+        )
+        extracted = extract_zip_upload(
+            session,
+            tenant_id=principal.tenant_id,
+            owner_type=owner_type,
+            owner_id=target_owner,
+            user_id=principal.user.id,
+            zip_data=zip_path,
+            folder=path,
+            campaign_id=campaign_id,
+            conflict_strategy=conflict_strategy,
+            conflict_resolutions=upload_resolutions,
+            metadata=metadata,
+            is_admin=_is_admin(principal),
+            encryption_vault_id=encryption_vault_id,
+            max_file_bytes=settings.file_upload_max_bytes,
+            max_total_bytes=settings.file_upload_zip_max_bytes,
+        )
+        _audit_connector_imports(session, principal, [item.asset for item in extracted])
+        session.commit()
+    except ConnectorPolicyDenied as exc:
+        session.rollback()
+        raise _connector_policy_error(exc) from exc
+    except (FileStorageError, UnsafeFilePathError, ValueError) as exc:
+        session.rollback()
+        raise _http_error(exc) from exc
+    finally:
+        if zip_path:
+            _cleanup_temp_file(zip_path)
+    return FileUploadResponse(
+        files=[
+            _asset_response(session, item.asset, include_shares=True)
+            for item in extracted
+        ]
+    )

src/govoplan_files/backend/schemas.py

@@ -1,5 +1,6 @@
 from __future__ import annotations
 
+from datetime import datetime
 from typing import Any, Literal
 
 from pydantic import BaseModel, Field
@@ -36,6 +37,7 @@ class FileConnectorSpaceCreateRequest(BaseModel):
     library_id: str | None = None
     remote_path: str = ""
     sync_mode: Literal["manual"] = "manual"
+    read_only: bool = True
     metadata: dict[str, Any] = Field(default_factory=dict)
 
 
@@ -44,6 +46,7 @@ class FileConnectorSpaceUpdateRequest(BaseModel):
     library_id: str | None = None
     remote_path: str | None = None
     sync_mode: Literal["manual"] | None = None
+    read_only: bool | None = None
     is_active: bool | None = None
     metadata: dict[str, Any] | None = None
 
@@ -73,11 +76,93 @@ class FileConnectorSpacesResponse(BaseModel):
 
 class FileShareResponse(BaseModel):
     id: str
+    file_asset_id: str
     target_type: str
     target_id: str
     permission: str
+    created_by_user_id: str | None = None
     created_at: str
+    expires_at: str | None = None
     revoked_at: str | None = None
+    revoked_by_user_id: str | None = None
+    active: bool
+
+
+class FileSharesResponse(BaseModel):
+    shares: list[FileShareResponse] = Field(default_factory=list)
+
+
+class FileIntegrityScanCreateRequest(BaseModel):
+    verify_checksums: bool = True
+    batch_size: int = Field(default=100, ge=1, le=1000)
+
+
+class FileIntegrityScanResponse(BaseModel):
+    id: str
+    tenant_id: str
+    storage_backend: str
+    storage_prefix: str
+    status: str
+    revision: int
+    phase: str
+    verify_checksums: bool
+    batch_size: int
+    scanned_blob_count: int
+    verified_blob_count: int
+    quarantined_blob_count: int
+    scanned_object_count: int
+    orphan_object_count: int
+    created_by_user_id: str | None = None
+    started_at: str | None = None
+    completed_at: str | None = None
+    last_error: str | None = None
+    created_at: str
+    updated_at: str
+
+
+class FileIntegrityScansResponse(BaseModel):
+    scans: list[FileIntegrityScanResponse] = Field(default_factory=list)
+
+
+class FileIntegrityFindingResponse(BaseModel):
+    id: str
+    scan_id: str
+    tenant_id: str
+    kind: str
+    state: str
+    revision: int
+    blob_id: str | None = None
+    storage_key: str
+    expected_size_bytes: int | None = None
+    observed_size_bytes: int | None = None
+    expected_checksum_sha256: str | None = None
+    observed_checksum_sha256: str | None = None
+    resolved_at: str | None = None
+    resolved_by_user_id: str | None = None
+    created_at: str
+    updated_at: str
+
+
+class FileIntegrityFindingsResponse(BaseModel):
+    findings: list[FileIntegrityFindingResponse] = Field(default_factory=list)
+
+
+class FileIntegrityActionRequest(BaseModel):
+    dry_run: bool = True
+    expected_revision: int = Field(ge=1)
+
+
+class FileIntegrityScanRunRequest(BaseModel):
+    expected_revision: int = Field(ge=1)
+
+
+class FileIntegrityActionResponse(BaseModel):
+    action: str
+    changed: bool
+    dry_run: bool
+    finding: FileIntegrityFindingResponse
+    inspection_kind: str | None = None
+    inspection_valid: bool | None = None
 
 
 class FileSourceProvenance(BaseModel):
@@ -109,6 +194,10 @@ class FileAssetResponse(BaseModel):
     created_at: str
     updated_at: str
     deleted_at: str | None = None
+    retained_until: str | None = None
+    legal_hold: bool = False
+    lifecycle_revision: int = 1
+    lifecycle_reason: str | None = None
     audit_relevant: bool = False
     metadata: dict[str, Any] | None = None
     source_provenance: FileSourceProvenance | None = None
@@ -154,6 +243,7 @@ class FileFolderDeleteResponse(BaseModel):
 
 class FileListResponse(BaseModel):
     files: list[FileAssetResponse]
+    total: int
     cursor: str | None = None
     next_cursor: str | None = None
     watermark: str | None = None
@@ -172,6 +262,27 @@ class FileUploadResponse(BaseModel):
     files: list[FileAssetResponse]
 
 
+class ArchiveEntryResponse(BaseModel):
+    path: str
+    kind: Literal["file", "directory"]
+    size_bytes: int
+    compressed_size_bytes: int | None = None
+    encrypted: bool = False
+
+
+class ArchivePreviewResponse(BaseModel):
+    preview_token: str
+    archive_format: str
+    entries: list[ArchiveEntryResponse] = Field(default_factory=list)
+    file_count: int
+    directory_count: int
+    expanded_size_bytes: int
+    compressed_size_bytes: int
+    requires_password: bool
+    password_verified: bool
+    expires_at: str
+
+
 class FileConnectorPolicySource(BaseModel):
     scope_type: Literal["system", "tenant", "user", "group", "campaign"] = "system"
     scope_id: str | None = None
@@ -372,6 +483,7 @@ class FileConnectorProviderResponse(BaseModel):
     installed: bool
     browse_supported: bool
     import_supported: bool
+    write_supported: bool = False
     optional_dependency: str | None = None
     permission_model: str
     sync_strategy: str
@@ -418,7 +530,7 @@ class FileConnectorImportRequest(BaseModel):
     target_folder: str | None = None
     target_path: str | None = None
     campaign_id: str | None = None
-    conflict_strategy: Literal["reject", "overwrite", "rename"] = "reject"
+    conflict_strategy: Literal["reject", "overwrite", "rename", "skip"] = "reject"
     source_revision: str | None = None
     metadata: dict[str, Any] = Field(default_factory=dict)
 
@@ -430,6 +542,77 @@ class FileConnectorSyncResponse(BaseModel):
     current_version_id: str
 
 
+class FileConnectorFolderSyncRequest(BaseModel):
+    path: str = ""
+    target_folder: str | None = None
+    recursive: bool = True
+    conflict_strategy: Literal["reject", "overwrite", "rename", "skip"] = "skip"
+    max_files: int = Field(default=100, ge=1, le=500)
+    max_depth: int = Field(default=12, ge=0, le=50)
+    metadata: dict[str, Any] = Field(default_factory=dict)
+
+
+class FileConnectorFolderSyncItemResponse(BaseModel):
+    source_path: str
+    target_path: str | None = None
+    action: Literal[
+        "created",
+        "updated",
+        "unchanged",
+        "skipped",
+        "conflict",
+        "policy_denied",
+        "failed",
+    ]
+    file: FileAssetResponse | None = None
+    previous_version_id: str | None = None
+    current_version_id: str | None = None
+    source_revision: str | None = None
+    detail: str | None = None
+    policy_decision: dict[str, Any] | None = None
+
+
+class FileConnectorFolderSyncSummary(BaseModel):
+    discovered: int = 0
+    created: int = 0
+    updated: int = 0
+    unchanged: int = 0
+    skipped: int = 0
+    conflicts: int = 0
+    policy_denied: int = 0
+    failed: int = 0
+
+
+class FileConnectorFolderSyncResponse(BaseModel):
+    connector_space_id: str
+    connector_profile_id: str
+    provider: str
+    remote_path: str
+    target_folder: str
+    recursive: bool
+    truncated: bool = False
+    summary: FileConnectorFolderSyncSummary
+    items: list[FileConnectorFolderSyncItemResponse] = Field(default_factory=list)
+
+
+class FileConnectorWriteRequest(BaseModel):
+    file_id: str
+    remote_path: str
+    idempotency_key: str = Field(min_length=8, max_length=160)
+    expected_revision: str | None = Field(default=None, max_length=500)
+
+
+class FileConnectorWriteResponse(BaseModel):
+    recovery_operation_id: str
+    status: str
+    replayed: bool = False
+    provider: str
+    remote_path: str
+    revision: str | None = None
+    checksum_sha256: str
+    size_bytes: int
+
+
 class BulkDeleteRequest(BaseModel):
     file_ids: list[str]
 
@@ -438,6 +621,69 @@ class BulkDeleteResponse(BaseModel):
     deleted_count: int
 
 
+class FileLifecycleUpdateRequest(BaseModel):
+    retained_until: datetime | None = None
+    legal_hold: bool = False
+    reason: str = Field(min_length=1, max_length=500)
+    expected_revision: int = Field(ge=1)
+
+
+class FileRestoreResponse(BaseModel):
+    restored_count: int
+
+
+class FileFolderRestoreResponse(BaseModel):
+    restored_folders: int
+    restored_files: int
+
+
+class FilePurgePreviewRequest(BaseModel):
+    file_ids: list[str] = Field(min_length=1, max_length=100)
+
+
+class FilePurgePreviewItem(BaseModel):
+    file_id: str
+    filename: str
+    lifecycle_revision: int
+    deleted_at: str | None = None
+    retained_until: str | None = None
+    legal_hold: bool = False
+    blockers: list[str] = Field(default_factory=list)
+    blob_ids: list[str] = Field(default_factory=list)
+
+
+class FilePurgePreviewResponse(BaseModel):
+    preview_sha256: str
+    eligible: bool
+    items: list[FilePurgePreviewItem]
+
+
+class FilePurgeExecuteRequest(FilePurgePreviewRequest):
+    preview_sha256: str = Field(min_length=64, max_length=64)
+    idempotency_key: str = Field(min_length=8, max_length=160)
+    approval_reference: str = Field(min_length=3, max_length=1000)
+    confirmation: Literal["PURGE"]
+
+
+class FilePurgeResponse(BaseModel):
+    recovery_operation_id: str
+    status: str
+    replayed: bool = False
+    purged_files: int = 0
+    released_blobs: int = 0
+
+
+class FileBlobGcRequest(BaseModel):
+    limit: int = Field(default=25, ge=1, le=100)
+    approval_reference: str = Field(min_length=3, max_length=1000)
+
+
+class FileBlobGcResponse(BaseModel):
+    inspected_blobs: int
+    deleted_blobs: int
+    unresolved_operation_ids: list[str] = Field(default_factory=list)
+
+
 class ConflictResolutionRequest(BaseModel):
     target_path: str
     action: Literal["overwrite", "rename", "skip"]
@@ -452,6 +698,7 @@ class FileShareRequest(BaseModel):
     target_type: Literal["user", "group", "campaign", "tenant"]
     target_id: str
     permission: Literal["read", "write", "manage"] = "read"
+    expires_at: datetime | None = None
 
 
 class BulkFileShareRequest(BaseModel):
@@ -459,6 +706,7 @@ class BulkFileShareRequest(BaseModel):
     target_type: Literal["user", "group", "campaign", "tenant"]
     target_id: str
     permission: Literal["read", "write", "manage"] = "read"
+    expires_at: datetime | None = None
 
 
 class BulkFileShareResponse(BaseModel):

src/govoplan_files/backend/search_source.py

@@ -0,0 +1,329 @@
+from __future__ import annotations
+
+from collections.abc import Mapping, Sequence
+from urllib.parse import quote
+
+from sqlalchemy import func, or_, select
+from sqlalchemy.orm import Session
+
+from govoplan_core.auth import ApiPrincipal
+from govoplan_core.core.events import PlatformEvent
+from govoplan_core.core.modules import ModuleContext
+from govoplan_core.core.search import (
+    SearchAuthorizationRequest,
+    SearchBackfillPage,
+    SearchBackfillRequest,
+    SearchDocument,
+    SearchIndexChange,
+    SearchResourceReference,
+    SearchResourceType,
+)
+from govoplan_files.backend.db.models import FileAsset, FileFolder, FileShare
+from govoplan_files.backend.storage.share_state import effective_file_share_clause
+
+
+PROVIDER_ID = "files.objects"
+RESOURCE_MODELS = {
+    "file": FileAsset,
+    "folder": FileFolder,
+}
+READ_SCOPE = "files:file:read"
+ADMIN_SCOPE = "files:file:admin"
+
+
+class FilesSearchSource:
+    def resource_types(self) -> Sequence[SearchResourceType]:
+        return (
+            SearchResourceType(
+                provider_id=PROVIDER_ID,
+                module_id="files",
+                resource_type="file",
+                label="Files",
+                requires_authorization_recheck=True,
+            ),
+            SearchResourceType(
+                provider_id=PROVIDER_ID,
+                module_id="files",
+                resource_type="folder",
+                label="File folders",
+                requires_authorization_recheck=True,
+            ),
+        )
+
+    def backfill(
+        self,
+        session: object,
+        *,
+        request: SearchBackfillRequest,
+    ) -> SearchBackfillPage:
+        db = _session(session)
+        model = _model(request.provider_id, request.resource_type)
+        statement = select(model).where(
+            model.tenant_id == request.tenant_id,
+            model.deleted_at.is_(None),
+        )
+        if request.cursor:
+            statement = statement.where(model.id > request.cursor)
+        rows = list(
+            db.scalars(
+                statement.order_by(model.id).limit(request.limit + 1)
+            )
+        )
+        has_more = len(rows) > request.limit
+        selected = rows[: request.limit]
+        shares = (
+            _shares_by_asset(db, selected)
+            if request.resource_type == "file"
+            else {}
+        )
+        high_watermark = db.scalar(
+            select(func.max(model.updated_at)).where(
+                model.tenant_id == request.tenant_id,
+                model.deleted_at.is_(None),
+            )
+        )
+        return SearchBackfillPage(
+            documents=tuple(
+                _document(
+                    row,
+                    resource_type=request.resource_type,
+                    shares=shares.get(row.id, ()),
+                )
+                for row in selected
+            ),
+            next_cursor=selected[-1].id if has_more and selected else None,
+            complete=not has_more,
+            high_watermark=(
+                high_watermark.isoformat()
+                if high_watermark is not None
+                else None
+            ),
+        )
+
+    def authorize(
+        self,
+        session: object,
+        principal: object,
+        *,
+        requests: Sequence[SearchAuthorizationRequest],
+    ) -> Mapping[str, bool]:
+        decisions = {item.reference.key: False for item in requests}
+        if not isinstance(principal, ApiPrincipal) or not (
+            principal.has(READ_SCOPE) or principal.has(ADMIN_SCOPE)
+        ):
+            return decisions
+        db = _session(session)
+        for request in requests:
+            reference = request.reference
+            if (
+                reference.tenant_id != principal.tenant_id
+                or reference.module_id != "files"
+                or reference.resource_type not in RESOURCE_MODELS
+            ):
+                continue
+            decisions[reference.key] = _can_read(
+                db,
+                principal,
+                reference=reference,
+            )
+        return decisions
+
+    def index_changes_for_event(
+        self,
+        session: object,
+        *,
+        event: PlatformEvent,
+        delivery_key: str,
+    ) -> Sequence[SearchIndexChange]:
+        if (
+            event.module_id != "files"
+            or event.tenant is None
+            or event.resource is None
+            or event.resource.id is None
+            or event.resource.type not in RESOURCE_MODELS
+        ):
+            return ()
+        db = _session(session)
+        reference = SearchResourceReference(
+            tenant_id=event.tenant.id,
+            module_id="files",
+            resource_type=event.resource.type,
+            resource_id=event.resource.id,
+        )
+        model = RESOURCE_MODELS[event.resource.type]
+        row = db.get(model, event.resource.id)
+        deleted = row is None or row.tenant_id != event.tenant.id or row.deleted_at is not None
+        cursor = event.event_id
+        document = None
+        if not deleted:
+            shares = (
+                tuple(_active_shares(db, row.id))
+                if event.resource.type == "file"
+                else ()
+            )
+            document = _document(
+                row,
+                resource_type=event.resource.type,
+                shares=shares,
+                change_cursor=cursor,
+            )
+        return (
+            SearchIndexChange(
+                change_id=f"{delivery_key}:{PROVIDER_ID}:{event.resource.type}",
+                provider_id=PROVIDER_ID,
+                kind="delete" if deleted else "upsert",
+                reference=reference,
+                source_revision=(
+                    document.source_revision if document is not None else cursor
+                ),
+                cursor=cursor,
+                document=document,
+                occurred_at=event.occurred_at,
+            ),
+        )
+
+
+def create_files_search_source(_context: ModuleContext) -> FilesSearchSource:
+    return FilesSearchSource()
+
+
+def _model(provider_id: str, resource_type: str):
+    if provider_id != PROVIDER_ID or resource_type not in RESOURCE_MODELS:
+        raise ValueError("Unsupported Files search source.")
+    return RESOURCE_MODELS[resource_type]
+
+
+def _document(
+    row: FileAsset | FileFolder,
+    *,
+    resource_type: str,
+    shares: Sequence[FileShare] = (),
+    change_cursor: str | None = None,
+) -> SearchDocument:
+    is_file = isinstance(row, FileAsset)
+    title = row.filename if is_file else (row.path.rsplit("/", 1)[-1] or row.path)
+    path = row.display_path if is_file else row.path
+    owner_id = row.owner_user_id if row.owner_type == "user" else row.owner_group_id
+    tokens = [f"scope:{READ_SCOPE}", f"scope:{ADMIN_SCOPE}"]
+    if owner_id:
+        tokens.append(
+            f"membership:{owner_id}"
+            if row.owner_type == "user"
+            else f"group:{owner_id}"
+        )
+    for share in shares:
+        prefix = "membership" if share.target_type == "user" else share.target_type
+        if prefix in {"membership", "group", "tenant"}:
+            tokens.append(f"{prefix}:{share.target_id}")
+    updated_at = row.updated_at or row.created_at
+    revision = (
+        f"{row.current_version_id or 'none'}:{updated_at.isoformat()}"
+        if is_file
+        else updated_at.isoformat()
+    )
+    return SearchDocument(
+        tenant_id=row.tenant_id,
+        module_id="files",
+        provider_id=PROVIDER_ID,
+        resource_type=resource_type,
+        resource_id=row.id,
+        title=title,
+        url=f"/files?{resource_type}Id={quote(row.id, safe='')}",
+        summary=((row.description or "") if is_file else path)[:4000] or None,
+        body=" ".join(
+            value for value in (path, row.description if is_file else None) if value
+        )[:200_000],
+        keywords=(path[:200], row.owner_type[:200]),
+        visibility="restricted",
+        acl_tokens=tuple(dict.fromkeys(tokens)),
+        metadata={
+            "path": path,
+            "owner_type": row.owner_type,
+            "current_version_id": row.current_version_id if is_file else None,
+        },
+        source_revision=revision,
+        change_cursor=change_cursor,
+        source_updated_at=updated_at,
+        requires_authorization_recheck=True,
+    )
+
+
+def _can_read(
+    session: Session,
+    principal: ApiPrincipal,
+    *,
+    reference: SearchResourceReference,
+) -> bool:
+    model = RESOURCE_MODELS[reference.resource_type]
+    row = session.get(model, reference.resource_id)
+    if row is None or row.tenant_id != principal.tenant_id or row.deleted_at is not None:
+        return False
+    if principal.has(ADMIN_SCOPE):
+        return True
+    user_id = str(getattr(principal.user, "id", "") or principal.membership_id or "")
+    if row.owner_type == "user" and row.owner_user_id == user_id:
+        return True
+    if row.owner_type == "group" and row.owner_group_id in principal.group_ids:
+        return True
+    if reference.resource_type != "file":
+        return False
+    target_clauses = [
+        (FileShare.target_type == "user") & (FileShare.target_id == user_id),
+        (FileShare.target_type == "tenant")
+        & (FileShare.target_id == principal.tenant_id),
+    ]
+    if principal.group_ids:
+        target_clauses.append(
+            (FileShare.target_type == "group")
+            & (FileShare.target_id.in_(tuple(principal.group_ids)))
+        )
+    return session.scalar(
+        select(FileShare.id).where(
+            FileShare.tenant_id == principal.tenant_id,
+            FileShare.file_asset_id == row.id,
+            effective_file_share_clause(),
+            or_(*target_clauses),
+        ).limit(1)
+    ) is not None
+
+
+def _active_shares(session: Session, asset_id: str) -> Sequence[FileShare]:
+    return tuple(
+        session.scalars(
+            select(FileShare).where(
+                FileShare.file_asset_id == asset_id,
+                effective_file_share_clause(),
+            )
+        )
+    )
+
+
+def _shares_by_asset(
+    session: Session,
+    rows: Sequence[FileAsset | FileFolder],
+) -> dict[str, tuple[FileShare, ...]]:
+    asset_ids = [row.id for row in rows if isinstance(row, FileAsset)]
+    grouped: dict[str, list[FileShare]] = {asset_id: [] for asset_id in asset_ids}
+    if not asset_ids:
+        return {}
+    for share in session.scalars(
+        select(FileShare).where(
+            FileShare.file_asset_id.in_(asset_ids),
+            effective_file_share_clause(),
+        )
+    ):
+        grouped.setdefault(share.file_asset_id, []).append(share)
+    return {key: tuple(value) for key, value in grouped.items()}
+
+
+def _session(value: object) -> Session:
+    if not isinstance(value, Session):
+        raise TypeError("Files search requires a SQLAlchemy session.")
+    return value
+
+
+__all__ = [
+    "FilesSearchSource",
+    "PROVIDER_ID",
+    "create_files_search_source",
+]

src/govoplan_files/backend/services/__init__.py

@@ -0,0 +1 @@
+"""Query and response assembly services used by Files routes."""

src/govoplan_files/backend/services/connector_settings_delta.py

@@ -0,0 +1,292 @@
+from __future__ import annotations
+
+from typing import Literal
+from sqlalchemy.orm import Session
+
+from govoplan_core.auth import ApiPrincipal
+from govoplan_core.api.v1.schemas import DeltaDeletedItem
+from govoplan_core.core.change_sequence import (
+    ChangeSequenceEntry,
+)
+from govoplan_files.backend.change_tracking import (
+    FILES_CONNECTOR_CREDENTIALS_COLLECTION,
+    FILES_CONNECTOR_POLICIES_COLLECTION,
+    FILES_CONNECTOR_PROFILES_COLLECTION,
+    FILES_CONNECTOR_SPACES_COLLECTION,
+)
+from govoplan_files.backend.schemas import (
+    FileConnectorCredentialResponse,
+    FileConnectorSettingsDeltaResponse,
+    FileConnectorPolicyResponse,
+    FileConnectorProfileResponse,
+)
+from govoplan_files.backend.db.models import FileConnectorSpace
+from govoplan_files.backend.storage.connector_credential_store import (
+    ConnectorCredential,
+)
+from govoplan_files.backend.storage.connector_profiles import ConnectorProfile
+from govoplan_files.backend.storage.connector_policy_store import (
+    connector_policy_response,
+)
+
+
+from govoplan_files.backend.route_support import (
+    FILES_CONNECTOR_CREDENTIAL_RESOURCE,
+    FILES_CONNECTOR_PROFILE_RESOURCE,
+    FILES_CONNECTOR_SPACE_RESOURCE,
+    _can_read_disabled_connector_profiles,
+    _connector_deleted_item,
+    _connector_space_response,
+    _ensure_campaign_file_access,
+    _file_connector_settings_response_watermark,
+    _file_connector_settings_watermark,
+    _visible_connector_credentials,
+    _visible_connector_profiles,
+    _visible_connector_spaces,
+)
+
+
+def _full_file_connector_settings_delta_response(
+    session: Session,
+    principal: ApiPrincipal,
+    *,
+    scope_type: str,
+    scope_id: str | None,
+    provider: str | None,
+    campaign_id: str | None,
+    include_disabled: bool,
+    include_inactive: bool,
+    owner_type: Literal["user", "group"] | None,
+    owner_id: str | None,
+) -> FileConnectorSettingsDeltaResponse:
+    if campaign_id:
+        _ensure_campaign_file_access(session, principal, campaign_id)
+    profiles = _visible_connector_profiles(
+        session,
+        principal,
+        provider=provider,
+        campaign_id=campaign_id,
+        include_disabled=include_disabled
+        and _can_read_disabled_connector_profiles(principal),
+        include_admin_scopes=_can_read_disabled_connector_profiles(principal),
+        include_effective_policy=False,
+    )
+    credentials = _visible_connector_credentials(
+        session,
+        principal,
+        provider=provider,
+        include_disabled=include_disabled,
+    )
+    spaces = _visible_connector_spaces(
+        session,
+        principal,
+        owner_type=owner_type,
+        owner_id=owner_id,
+        include_inactive=include_inactive,
+    )
+    return FileConnectorSettingsDeltaResponse(
+        profiles=[
+            FileConnectorProfileResponse(**profile.to_response())
+            for profile in profiles
+        ],
+        credentials=[
+            FileConnectorCredentialResponse(**credential.to_response())
+            for credential in credentials
+        ],
+        spaces=[_connector_space_response(space) for space in spaces],
+        policy=FileConnectorPolicyResponse(
+            **connector_policy_response(
+                session,
+                tenant_id=principal.tenant_id,
+                scope_type=scope_type,
+                scope_id=scope_id,
+            )
+        ),
+        changed_sections=["profiles", "credentials", "spaces", "policy"],
+        deleted=[],
+        watermark=_file_connector_settings_watermark(
+            session, tenant_id=principal.tenant_id
+        ),
+        has_more=False,
+        full=True,
+    )
+
+
+def _changed_file_connector_setting_ids(
+    entries: list[ChangeSequenceEntry],
+) -> tuple[set[str], set[str], set[str], bool]:
+    changed_profile_ids = {
+        entry.resource_id
+        for entry in entries
+        if entry.collection == FILES_CONNECTOR_PROFILES_COLLECTION
+        and entry.resource_type == FILES_CONNECTOR_PROFILE_RESOURCE
+    }
+    changed_credential_ids = {
+        entry.resource_id
+        for entry in entries
+        if entry.collection == FILES_CONNECTOR_CREDENTIALS_COLLECTION
+        and entry.resource_type == FILES_CONNECTOR_CREDENTIAL_RESOURCE
+    }
+    changed_space_ids = {
+        entry.resource_id
+        for entry in entries
+        if entry.collection == FILES_CONNECTOR_SPACES_COLLECTION
+        and entry.resource_type == FILES_CONNECTOR_SPACE_RESOURCE
+    }
+    policy_changed = any(
+        entry.collection == FILES_CONNECTOR_POLICIES_COLLECTION for entry in entries
+    )
+    return (
+        changed_profile_ids,
+        changed_credential_ids,
+        changed_space_ids,
+        policy_changed,
+    )
+
+
+def _file_connector_settings_changed_sections(
+    *,
+    changed_profile_ids: set[str],
+    changed_credential_ids: set[str],
+    changed_space_ids: set[str],
+    policy_changed: bool,
+    profiles: list[ConnectorProfile],
+) -> list[str]:
+    changed_sections = []
+    if changed_profile_ids or any(
+        profile.credential_profile_id
+        and profile.credential_profile_id in changed_credential_ids
+        for profile in profiles
+    ):
+        changed_sections.append("profiles")
+    if changed_credential_ids:
+        changed_sections.append("credentials")
+    if changed_space_ids:
+        changed_sections.append("spaces")
+    if policy_changed:
+        changed_sections.append("policy")
+    return changed_sections
+
+
+def _file_connector_settings_deleted_items(
+    entries: list[ChangeSequenceEntry],
+    *,
+    visible_profiles: dict[str, ConnectorProfile],
+    visible_credentials: dict[str, ConnectorCredential],
+    visible_spaces: dict[str, FileConnectorSpace],
+) -> list[DeltaDeletedItem]:
+    return [
+        _connector_deleted_item(entry)
+        for entry in entries
+        if (
+            entry.resource_type == FILES_CONNECTOR_PROFILE_RESOURCE
+            and entry.resource_id not in visible_profiles
+        )
+        or (
+            entry.resource_type == FILES_CONNECTOR_CREDENTIAL_RESOURCE
+            and entry.resource_id not in visible_credentials
+        )
+        or (
+            entry.resource_type == FILES_CONNECTOR_SPACE_RESOURCE
+            and entry.resource_id not in visible_spaces
+        )
+    ]
+
+
+def _incremental_file_connector_settings_delta_response(
+    session: Session,
+    principal: ApiPrincipal,
+    *,
+    entries: list[ChangeSequenceEntry],
+    has_more: bool,
+    scope_type: str,
+    scope_id: str | None,
+    provider: str | None,
+    campaign_id: str | None,
+    include_disabled: bool,
+    include_inactive: bool,
+    owner_type: Literal["user", "group"] | None,
+    owner_id: str | None,
+) -> FileConnectorSettingsDeltaResponse:
+    changed_profile_ids, changed_credential_ids, changed_space_ids, policy_changed = (
+        _changed_file_connector_setting_ids(entries)
+    )
+    profiles = _visible_connector_profiles(
+        session,
+        principal,
+        provider=provider,
+        campaign_id=campaign_id,
+        include_disabled=include_disabled
+        and _can_read_disabled_connector_profiles(principal),
+        include_admin_scopes=_can_read_disabled_connector_profiles(principal),
+        include_effective_policy=False,
+    )
+    visible_profiles = {
+        profile.id: profile
+        for profile in profiles
+        if profile.id in changed_profile_ids
+        or (
+            profile.credential_profile_id
+            and profile.credential_profile_id in changed_credential_ids
+        )
+    }
+    credentials = _visible_connector_credentials(
+        session,
+        principal,
+        provider=provider,
+        include_disabled=include_disabled,
+    )
+    visible_credentials = {
+        credential.id: credential
+        for credential in credentials
+        if credential.id in changed_credential_ids
+    }
+    spaces = _visible_connector_spaces(
+        session,
+        principal,
+        owner_type=owner_type,
+        owner_id=owner_id,
+        include_inactive=include_inactive,
+    )
+    visible_spaces = {
+        space.id: space for space in spaces if space.id in changed_space_ids
+    }
+    return FileConnectorSettingsDeltaResponse(
+        profiles=[
+            FileConnectorProfileResponse(**profile.to_response())
+            for profile in visible_profiles.values()
+        ],
+        credentials=[
+            FileConnectorCredentialResponse(**credential.to_response())
+            for credential in visible_credentials.values()
+        ],
+        spaces=[_connector_space_response(space) for space in visible_spaces.values()],
+        policy=FileConnectorPolicyResponse(
+            **connector_policy_response(
+                session,
+                tenant_id=principal.tenant_id,
+                scope_type=scope_type,
+                scope_id=scope_id,
+            )
+        )
+        if policy_changed
+        else None,
+        changed_sections=_file_connector_settings_changed_sections(
+            changed_profile_ids=changed_profile_ids,
+            changed_credential_ids=changed_credential_ids,
+            changed_space_ids=changed_space_ids,
+            policy_changed=policy_changed,
+            profiles=profiles,
+        ),
+        deleted=_file_connector_settings_deleted_items(
+            entries,
+            visible_profiles=visible_profiles,
+            visible_credentials=visible_credentials,
+            visible_spaces=visible_spaces,
+        ),
+        watermark=_file_connector_settings_response_watermark(
+            session, tenant_id=principal.tenant_id, entries=entries, has_more=has_more
+        ),
+        has_more=has_more,
+        full=False,
+    )

src/govoplan_files/backend/services/list_queries.py

@@ -0,0 +1,620 @@
+from __future__ import annotations
+
+from datetime import datetime
+from typing import Literal
+from fastapi import HTTPException, status
+from sqlalchemy.orm import Session
+
+from govoplan_core.auth import ApiPrincipal
+from govoplan_core.api.v1.schemas import DeltaDeletedItem
+from govoplan_core.core.change_sequence import (
+    ChangeSequenceEntry,
+    decode_sequence_watermark,
+    encode_sequence_watermark,
+    max_sequence_id,
+    sequence_entries_since,
+    sequence_watermark_is_expired,
+)
+from govoplan_core.core.pagination import (
+    KeysetCursorError,
+    decode_keyset_cursor,
+    encode_keyset_cursor,
+    keyset_query_fingerprint,
+)
+from govoplan_files.backend.change_tracking import (
+    FILES_ASSETS_COLLECTION,
+    FILES_FOLDERS_COLLECTION,
+    FILES_MODULE_ID,
+)
+from govoplan_files.backend.schemas import (
+    FileDeltaResponse,
+)
+from govoplan_files.backend.db.models import FileAsset, FileFolder, FileShare
+from govoplan_files.backend.storage.access import user_group_ids
+from govoplan_files.backend.storage.files import (
+    list_assets_for_user,
+)
+from govoplan_files.backend.storage.folders import list_folders_for_user
+from govoplan_files.backend.storage.share_state import effective_file_share_clause
+
+
+from govoplan_files.backend.route_support import (
+    _asset_list_response,
+    _folder_response,
+    _is_admin,
+)
+
+_FILES_DELTA_COLLECTIONS = (FILES_ASSETS_COLLECTION, FILES_FOLDERS_COLLECTION)
+FILES_LIST_CURSOR_SCOPE = "files.list.v1"
+FOLDERS_LIST_CURSOR_SCOPE = "files.folders.list.v1"
+DEFAULT_FILE_LIST_PAGE_SIZE = 500
+
+
+def _cursor_http_error(exc: Exception) -> HTTPException:
+    return HTTPException(status_code=status.HTTP_400_BAD_REQUEST, detail=str(exc))
+
+
+def _cursor_page_size(
+    scope: str, cursor: str | None, explicit_page_size: int | None
+) -> int | None:
+    if explicit_page_size is not None:
+        return explicit_page_size
+    if not cursor:
+        return DEFAULT_FILE_LIST_PAGE_SIZE
+    try:
+        values = decode_keyset_cursor(scope, cursor)
+    except KeysetCursorError as exc:
+        raise _cursor_http_error(exc) from exc
+    raw_page_size = values.get("page_size")
+    if not isinstance(raw_page_size, int) or raw_page_size < 1 or raw_page_size > 1000:
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST, detail="Invalid pagination cursor"
+        )
+    return raw_page_size
+
+
+def _files_list_fingerprint(
+    principal: ApiPrincipal,
+    *,
+    owner_type: Literal["user", "group"] | None,
+    owner_id: str | None,
+    campaign_id: str | None,
+    path_prefix: str | None,
+    campaign_usage: Literal["linked", "unlinked"] | None,
+    audit_relevant: bool | None,
+    page_size: int,
+) -> str:
+    return keyset_query_fingerprint(
+        FILES_LIST_CURSOR_SCOPE,
+        {
+            "tenant_id": principal.tenant_id,
+            "actor": "admin" if _is_admin(principal) else principal.user.id,
+            "owner_type": owner_type,
+            "owner_id": owner_id,
+            "campaign_id": campaign_id,
+            "path_prefix": path_prefix or "",
+            "campaign_usage": campaign_usage or "",
+            "audit_relevant": audit_relevant,
+            "sort": "display_path.asc,updated_at.desc,id.asc",
+            "page_size": page_size,
+        },
+    )
+
+
+def _folders_list_fingerprint(
+    principal: ApiPrincipal,
+    *,
+    owner_type: Literal["user", "group"],
+    owner_id: str,
+    page_size: int,
+) -> str:
+    return keyset_query_fingerprint(
+        FOLDERS_LIST_CURSOR_SCOPE,
+        {
+            "tenant_id": principal.tenant_id,
+            "actor": "admin" if _is_admin(principal) else principal.user.id,
+            "owner_type": owner_type,
+            "owner_id": owner_id,
+            "sort": "path.asc,id.asc",
+            "page_size": page_size,
+        },
+    )
+
+
+def _file_cursor_values(
+    cursor: str | None, *, fingerprint: str
+) -> tuple[str | None, datetime | None, str | None]:
+    try:
+        values = decode_keyset_cursor(
+            FILES_LIST_CURSOR_SCOPE, cursor, fingerprint=fingerprint
+        )
+    except KeysetCursorError as exc:
+        raise _cursor_http_error(exc) from exc
+    if values is None:
+        return None, None, None
+    display_path = values.get("display_path")
+    updated_at = values.get("updated_at")
+    asset_id = values.get("id")
+    if (
+        not isinstance(display_path, str)
+        or not isinstance(updated_at, str)
+        or not isinstance(asset_id, str)
+    ):
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST, detail="Invalid pagination cursor"
+        )
+    try:
+        parsed_updated_at = datetime.fromisoformat(updated_at)
+    except ValueError as exc:
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST, detail="Invalid pagination cursor"
+        ) from exc
+    return display_path, parsed_updated_at, asset_id
+
+
+def _folder_cursor_values(
+    cursor: str | None, *, fingerprint: str
+) -> tuple[str | None, str | None]:
+    try:
+        values = decode_keyset_cursor(
+            FOLDERS_LIST_CURSOR_SCOPE, cursor, fingerprint=fingerprint
+        )
+    except KeysetCursorError as exc:
+        raise _cursor_http_error(exc) from exc
+    if values is None:
+        return None, None
+    folder_path = values.get("path")
+    folder_id = values.get("id")
+    if not isinstance(folder_path, str) or not isinstance(folder_id, str):
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST, detail="Invalid pagination cursor"
+        )
+    return folder_path, folder_id
+
+
+def _next_file_list_cursor(
+    principal: ApiPrincipal,
+    assets: list[FileAsset],
+    *,
+    owner_type: Literal["user", "group"] | None,
+    owner_id: str | None,
+    campaign_id: str | None,
+    path_prefix: str | None,
+    campaign_usage: Literal["linked", "unlinked"] | None,
+    audit_relevant: bool | None,
+    page_size: int,
+    has_more: bool,
+) -> str | None:
+    if not has_more or not assets:
+        return None
+    last = assets[-1]
+    return encode_keyset_cursor(
+        FILES_LIST_CURSOR_SCOPE,
+        fingerprint=_files_list_fingerprint(
+            principal,
+            owner_type=owner_type,
+            owner_id=owner_id,
+            campaign_id=campaign_id,
+            path_prefix=path_prefix,
+            campaign_usage=campaign_usage,
+            audit_relevant=audit_relevant,
+            page_size=page_size,
+        ),
+        values={
+            "display_path": last.display_path,
+            "updated_at": last.updated_at,
+            "id": last.id,
+            "page_size": page_size,
+        },
+    )
+
+
+def _next_folder_list_cursor(
+    principal: ApiPrincipal,
+    folders: list[FileFolder],
+    *,
+    owner_type: Literal["user", "group"],
+    owner_id: str,
+    page_size: int,
+    has_more: bool,
+) -> str | None:
+    if not has_more or not folders:
+        return None
+    last = folders[-1]
+    return encode_keyset_cursor(
+        FOLDERS_LIST_CURSOR_SCOPE,
+        fingerprint=_folders_list_fingerprint(
+            principal, owner_type=owner_type, owner_id=owner_id, page_size=page_size
+        ),
+        values={"path": last.path, "id": last.id, "page_size": page_size},
+    )
+
+
+def _files_delta_watermark(session: Session, tenant_id: str) -> str:
+    return encode_sequence_watermark(
+        max_sequence_id(
+            session,
+            tenant_id=tenant_id,
+            module_id=FILES_MODULE_ID,
+            collections=_FILES_DELTA_COLLECTIONS,
+        )
+    )
+
+
+def _full_file_delta_response(
+    session: Session,
+    *,
+    principal: ApiPrincipal,
+    owner_type: Literal["user", "group"] | None,
+    owner_id: str | None,
+    campaign_id: str | None,
+    path_prefix: str | None,
+) -> FileDeltaResponse:
+    assets = list_assets_for_user(
+        session,
+        tenant_id=principal.tenant_id,
+        user_id=principal.user.id,
+        owner_type=owner_type,
+        owner_id=owner_id,
+        campaign_id=campaign_id,
+        path_prefix=path_prefix,
+        is_admin=_is_admin(principal),
+    )
+    folders = _visible_folders_for_delta(
+        session,
+        principal=principal,
+        owner_type=owner_type,
+        owner_id=owner_id,
+        path_prefix=path_prefix,
+    )
+    return FileDeltaResponse(
+        files=_asset_list_response(session, assets, include_shares=True),
+        folders=[_folder_response(folder) for folder in folders],
+        deleted=[],
+        watermark=_files_delta_watermark(session, principal.tenant_id),
+        has_more=False,
+        full=True,
+    )
+
+
+def _visible_folders_for_delta(
+    session: Session,
+    *,
+    principal: ApiPrincipal,
+    owner_type: Literal["user", "group"] | None,
+    owner_id: str | None,
+    path_prefix: str | None,
+) -> list[FileFolder]:
+    if not owner_type or not owner_id:
+        return []
+    folders = list_folders_for_user(
+        session,
+        tenant_id=principal.tenant_id,
+        user_id=principal.user.id,
+        owner_type=owner_type,
+        owner_id=owner_id,
+        is_admin=_is_admin(principal),
+    )
+    if not path_prefix:
+        return folders
+    normalized = path_prefix.strip().strip("/")
+    if not normalized:
+        return folders
+    return [
+        folder
+        for folder in folders
+        if folder.path == normalized or folder.path.startswith(f"{normalized}/")
+    ]
+
+
+def _entry_path_matches(
+    entry_payload: dict[str, object], path_prefix: str | None
+) -> bool:
+    if not path_prefix:
+        return True
+    normalized = path_prefix.strip().strip("/")
+    if not normalized:
+        return True
+    for key in ("path", "previous_path"):
+        value = entry_payload.get(key)
+        if isinstance(value, str) and (
+            value == normalized or value.startswith(f"{normalized}/")
+        ):
+            return True
+    return False
+
+
+def _entry_owner_matches(
+    entry_payload: dict[str, object],
+    owner_type: Literal["user", "group"] | None,
+    owner_id: str | None,
+) -> bool:
+    if not owner_type or not owner_id:
+        return True
+    return (
+        entry_payload.get("owner_type") == owner_type
+        and entry_payload.get("owner_id") == owner_id
+    )
+
+
+def _entry_campaign_matches(session: Session, entry, campaign_id: str | None) -> bool:
+    if not campaign_id:
+        return True
+    payload = entry.payload or {}
+    if (
+        payload.get("share_target_type") == "campaign"
+        and payload.get("share_target_id") == campaign_id
+    ):
+        return True
+    if entry.resource_type != "file":
+        return False
+    return (
+        session.query(FileShare)
+        .filter(
+            FileShare.tenant_id == entry.tenant_id,
+            FileShare.file_asset_id == entry.resource_id,
+            FileShare.target_type == "campaign",
+            FileShare.target_id == campaign_id,
+            effective_file_share_clause(),
+        )
+        .first()
+        is not None
+    )
+
+
+def _principal_group_ids_for_delta(
+    session: Session, principal: ApiPrincipal
+) -> set[str]:
+    cache = session.info.setdefault("files_delta_group_ids", {})
+    key = (principal.tenant_id, principal.user.id)
+    if key not in cache:
+        cache[key] = set(
+            user_group_ids(
+                session, tenant_id=principal.tenant_id, user_id=principal.user.id
+            )
+        )
+    return cache[key]
+
+
+def _entry_subject_matches_principal(
+    session: Session, principal: ApiPrincipal, entry_payload: dict[str, object]
+) -> bool:
+    if _is_admin(principal):
+        return True
+    owner_type = entry_payload.get("owner_type")
+    owner_id = entry_payload.get("owner_id")
+    if owner_type == "user" and owner_id == principal.user.id:
+        return True
+    if owner_type == "group" and isinstance(owner_id, str):
+        if owner_id in _principal_group_ids_for_delta(session, principal):
+            return True
+    share_target_type = entry_payload.get("share_target_type")
+    share_target_id = entry_payload.get("share_target_id")
+    if share_target_type == "user" and share_target_id == principal.user.id:
+        return True
+    if share_target_type == "tenant" and share_target_id == principal.tenant_id:
+        return True
+    if share_target_type == "group" and isinstance(share_target_id, str):
+        if share_target_id in _principal_group_ids_for_delta(session, principal):
+            return True
+    return False
+
+
+def _entry_matches_delta_scope(
+    session: Session,
+    principal: ApiPrincipal,
+    entry,
+    *,
+    owner_type: Literal["user", "group"] | None,
+    owner_id: str | None,
+    campaign_id: str | None,
+    path_prefix: str | None,
+) -> bool:
+    payload = entry.payload or {}
+    if (
+        not owner_type
+        and not campaign_id
+        and not _entry_subject_matches_principal(session, principal, payload)
+    ):
+        return False
+    return (
+        _entry_owner_matches(payload, owner_type, owner_id)
+        and _entry_path_matches(payload, path_prefix)
+        and _entry_campaign_matches(session, entry, campaign_id)
+    )
+
+
+def _decode_files_delta_watermark(since: str) -> int:
+    try:
+        return decode_sequence_watermark(since)
+    except ValueError as exc:
+        raise HTTPException(
+            status_code=status.HTTP_400_BAD_REQUEST, detail=str(exc)
+        ) from exc
+
+
+def _changed_delta_resource_ids(
+    entries: list[ChangeSequenceEntry],
+) -> tuple[list[str], list[str]]:
+    file_ids = list(
+        dict.fromkeys(
+            entry.resource_id for entry in entries if entry.resource_type == "file"
+        )
+    )
+    folder_ids = list(
+        dict.fromkeys(
+            entry.resource_id for entry in entries if entry.resource_type == "folder"
+        )
+    )
+    return file_ids, folder_ids
+
+
+def _visible_assets_for_delta(
+    session: Session,
+    *,
+    principal: ApiPrincipal,
+    owner_type: Literal["user", "group"] | None,
+    owner_id: str | None,
+    campaign_id: str | None,
+    path_prefix: str | None,
+    changed_file_ids: list[str],
+) -> dict[str, FileAsset]:
+    return {
+        asset.id: asset
+        for asset in list_assets_for_user(
+            session,
+            tenant_id=principal.tenant_id,
+            user_id=principal.user.id,
+            owner_type=owner_type,
+            owner_id=owner_id,
+            campaign_id=campaign_id,
+            path_prefix=path_prefix,
+            is_admin=_is_admin(principal),
+        )
+        if asset.id in changed_file_ids
+    }
+
+
+def _changed_visible_folders_for_delta(
+    session: Session,
+    *,
+    principal: ApiPrincipal,
+    owner_type: Literal["user", "group"] | None,
+    owner_id: str | None,
+    path_prefix: str | None,
+    changed_folder_ids: list[str],
+) -> dict[str, FileFolder]:
+    return {
+        folder.id: folder
+        for folder in _visible_folders_for_delta(
+            session,
+            principal=principal,
+            owner_type=owner_type,
+            owner_id=owner_id,
+            path_prefix=path_prefix,
+        )
+        if folder.id in changed_folder_ids
+    }
+
+
+def _deleted_delta_items(
+    session: Session,
+    *,
+    principal: ApiPrincipal,
+    entries: list[ChangeSequenceEntry],
+    visible_assets: dict[str, FileAsset],
+    visible_folders: dict[str, FileFolder],
+    owner_type: Literal["user", "group"] | None,
+    owner_id: str | None,
+    campaign_id: str | None,
+    path_prefix: str | None,
+) -> list[DeltaDeletedItem]:
+    deleted: dict[tuple[str, str], DeltaDeletedItem] = {}
+    for entry in entries:
+        if entry.resource_type == "file" and entry.resource_id in visible_assets:
+            continue
+        if entry.resource_type == "folder" and entry.resource_id in visible_folders:
+            continue
+        if not _entry_matches_delta_scope(
+            session,
+            principal,
+            entry,
+            owner_type=owner_type,
+            owner_id=owner_id,
+            campaign_id=campaign_id,
+            path_prefix=path_prefix,
+        ):
+            continue
+        deleted[(entry.resource_type, entry.resource_id)] = DeltaDeletedItem(
+            id=entry.resource_id,
+            resource_type=entry.resource_type,
+            revision=encode_sequence_watermark(entry.id),
+            deleted_at=entry.created_at if entry.operation == "deleted" else None,
+        )
+    return list(deleted.values())
+
+
+def _files_delta_response(
+    session: Session,
+    *,
+    principal: ApiPrincipal,
+    owner_type: Literal["user", "group"] | None,
+    owner_id: str | None,
+    campaign_id: str | None,
+    path_prefix: str | None,
+    since: str,
+    limit: int,
+) -> FileDeltaResponse:
+    since_sequence = _decode_files_delta_watermark(since)
+    if sequence_watermark_is_expired(
+        session,
+        since=since_sequence,
+        tenant_id=principal.tenant_id,
+        module_id=FILES_MODULE_ID,
+        collections=_FILES_DELTA_COLLECTIONS,
+    ):
+        return _full_file_delta_response(
+            session,
+            principal=principal,
+            owner_type=owner_type,
+            owner_id=owner_id,
+            campaign_id=campaign_id,
+            path_prefix=path_prefix,
+        )
+
+    entries_plus_one = sequence_entries_since(
+        session,
+        since=since_sequence,
+        tenant_id=principal.tenant_id,
+        module_id=FILES_MODULE_ID,
+        collections=_FILES_DELTA_COLLECTIONS,
+        limit=limit + 1,
+    )
+    has_more = len(entries_plus_one) > limit
+    entries = entries_plus_one[:limit]
+
+    changed_file_ids, changed_folder_ids = _changed_delta_resource_ids(entries)
+    visible_assets = _visible_assets_for_delta(
+        session,
+        principal=principal,
+        owner_type=owner_type,
+        owner_id=owner_id,
+        campaign_id=campaign_id,
+        path_prefix=path_prefix,
+        changed_file_ids=changed_file_ids,
+    )
+    visible_folders = _changed_visible_folders_for_delta(
+        session,
+        principal=principal,
+        owner_type=owner_type,
+        owner_id=owner_id,
+        path_prefix=path_prefix,
+        changed_folder_ids=changed_folder_ids,
+    )
+    deleted = _deleted_delta_items(
+        session,
+        principal=principal,
+        entries=entries,
+        visible_assets=visible_assets,
+        visible_folders=visible_folders,
+        owner_type=owner_type,
+        owner_id=owner_id,
+        campaign_id=campaign_id,
+        path_prefix=path_prefix,
+    )
+
+    watermark = (
+        encode_sequence_watermark(entries[-1].id)
+        if has_more and entries
+        else _files_delta_watermark(session, principal.tenant_id)
+    )
+    return FileDeltaResponse(
+        files=_asset_list_response(
+            session, list(visible_assets.values()), include_shares=True
+        ),
+        folders=[_folder_response(folder) for folder in visible_folders.values()],
+        deleted=deleted,
+        watermark=watermark,
+        has_more=has_more,
+        full=False,
+    )

src/govoplan_files/backend/storage/archives.py

@@ -1,55 +1,91 @@
 from __future__ import annotations
 
 import mimetypes
+import re
+import stat
+import tarfile
 import zipfile
+from dataclasses import dataclass
 from io import BytesIO
 from os import PathLike
-from pathlib import Path
+from pathlib import Path, PurePosixPath
-from typing import Any, Iterable
+from typing import Any, BinaryIO, Iterable, Literal
 
+import pyzipper
 from sqlalchemy.orm import Session
 
 from govoplan_files.backend.db.models import FileAsset
-from govoplan_files.backend.storage.common import FileConflictResolution, FileStorageError, UploadedStoredFile
+from govoplan_files.backend.storage.backends import (
-from govoplan_files.backend.storage.backends import StorageBackendError, get_storage_backend
+    StorageBackendError,
-from govoplan_files.backend.storage.files import create_file_asset, current_versions_and_blobs
+    get_storage_backend,
-from govoplan_files.backend.storage.paths import filename_from_path, normalize_folder, normalize_logical_path
+)
+from govoplan_files.backend.storage.common import (
+    FileConflictResolution,
+    FileStorageError,
+    UploadedStoredFile,
+)
+from govoplan_files.backend.storage.files import (
+    create_file_asset,
+    current_versions_and_blobs,
+)
+from govoplan_files.backend.storage.paths import (
+    filename_from_path,
+    normalize_folder,
+    normalize_logical_path,
+)
 
 
-_ZIP_READ_CHUNK_SIZE = 1024 * 1024
+_ARCHIVE_READ_CHUNK_SIZE = 1024 * 1024
-ZIP_UPLOAD_MAX_FILES = 1000
+_WINDOWS_DRIVE_RE = re.compile(r"^[A-Za-z]:")
+ARCHIVE_UPLOAD_MAX_ENTRIES = 10_000
+# Kept for callers and documentation using the former ZIP-specific name.
+ZIP_UPLOAD_MAX_FILES = ARCHIVE_UPLOAD_MAX_ENTRIES
+SUPPORTED_ARCHIVE_SUFFIXES = (
+    ".tar.bz2",
+    ".tar.gz",
+    ".tar.xz",
+    ".tbz2",
+    ".tgz",
+    ".txz",
+    ".tar",
+    ".zip",
+)
 
 
-def _read_zip_member(
+class ArchivePasswordError(FileStorageError):
-    archive: zipfile.ZipFile,
+    pass
-    info: zipfile.ZipInfo,
-    *,
-    max_file_bytes: int,
-    max_total_bytes: int,
-    current_total: int,
-) -> tuple[bytes, int]:
-    parts: list[bytes] = []
-    actual_size = 0
-    with archive.open(info) as source:
-        while True:
-            read_size = min(_ZIP_READ_CHUNK_SIZE, max_file_bytes + 1 - actual_size)
-            chunk = source.read(read_size)
-            if not chunk:
-                break
-            actual_size += len(chunk)
-            if actual_size > max_file_bytes:
-                raise FileStorageError(f"ZIP member {info.filename!r} exceeds per-file limit")
-            if current_total + actual_size > max_total_bytes:
-                raise FileStorageError("ZIP is too large after extraction")
-            parts.append(chunk)
-    return b"".join(parts), current_total + actual_size
 
 
-def create_zip_file(session: Session, assets: Iterable[FileAsset], output_path: str | Path) -> None:
+@dataclass(frozen=True, slots=True)
+class ArchiveEntry:
+    path: str
+    kind: Literal["file", "directory"]
+    size_bytes: int
+    compressed_size_bytes: int | None = None
+    encrypted: bool = False
+
+
+@dataclass(frozen=True, slots=True)
+class ArchiveInspection:
+    archive_format: str
+    entries: tuple[ArchiveEntry, ...]
+    file_count: int
+    directory_count: int
+    expanded_size_bytes: int
+    compressed_size_bytes: int
+    requires_password: bool
+    password_verified: bool
+
+
+def create_zip_file(
+    session: Session, assets: Iterable[FileAsset], output_path: str | Path
+) -> None:
     backend = get_storage_backend()
     asset_list = list(assets)
     version_blobs = current_versions_and_blobs(session, asset_list)
-    with zipfile.ZipFile(output_path, mode="w", compression=zipfile.ZIP_DEFLATED) as archive:
+    with zipfile.ZipFile(
+        output_path, mode="w", compression=zipfile.ZIP_DEFLATED
+    ) as archive:
         for asset in asset_list:
             _version, blob = version_blobs[asset.id]
             info = zipfile.ZipInfo(asset.display_path)
@@ -63,6 +99,147 @@ def create_zip_file(session: Session, assets: Iterable[FileAsset], output_path:
                     raise FileStorageError(str(exc)) from exc
 
 
+def archive_format_for_filename(filename: str) -> str:
+    lowered = filename.strip().casefold()
+    if lowered.endswith(".zip"):
+        return "zip"
+    if lowered.endswith((".tar.gz", ".tgz")):
+        return "tar.gz"
+    if lowered.endswith((".tar.bz2", ".tbz2")):
+        return "tar.bz2"
+    if lowered.endswith((".tar.xz", ".txz")):
+        return "tar.xz"
+    if lowered.endswith(".tar"):
+        return "tar"
+    raise FileStorageError(
+        "Unsupported archive format. Use ZIP, TAR, TAR.GZ, TAR.BZ2, or TAR.XZ."
+    )
+
+
+def is_supported_archive_filename(filename: str) -> bool:
+    try:
+        archive_format_for_filename(filename)
+    except FileStorageError:
+        return False
+    return True
+
+
+def inspect_archive(
+    archive_data: bytes | str | PathLike[str],
+    *,
+    filename: str,
+    password: str | None = None,
+    max_entries: int = ARCHIVE_UPLOAD_MAX_ENTRIES,
+    max_expanded_bytes: int = 2 * 1024 * 1024 * 1024,
+    max_expansion_ratio: int = 100,
+) -> ArchiveInspection:
+    archive_format = archive_format_for_filename(filename)
+    compressed_size = _archive_size(archive_data)
+    if archive_format == "zip":
+        entries, requires_password, password_verified = _inspect_zip(
+            archive_data,
+            password=password,
+        )
+    else:
+        entries = _inspect_tar(archive_data)
+        requires_password = False
+        password_verified = True
+    _validate_archive_limits(
+        entries,
+        compressed_size=compressed_size,
+        max_entries=max_entries,
+        max_expanded_bytes=max_expanded_bytes,
+        max_expansion_ratio=max_expansion_ratio,
+    )
+    complete_entries = _with_derived_directories(entries)
+    return ArchiveInspection(
+        archive_format=archive_format,
+        entries=tuple(complete_entries),
+        file_count=sum(entry.kind == "file" for entry in complete_entries),
+        directory_count=sum(
+            entry.kind == "directory" for entry in complete_entries
+        ),
+        expanded_size_bytes=sum(
+            entry.size_bytes for entry in entries if entry.kind == "file"
+        ),
+        compressed_size_bytes=compressed_size,
+        requires_password=requires_password,
+        password_verified=password_verified,
+    )
+
+
+def extract_archive_upload(
+    session: Session,
+    *,
+    tenant_id: str,
+    owner_type: str,
+    owner_id: str,
+    user_id: str,
+    archive_data: bytes | str | PathLike[str],
+    filename: str,
+    folder: str | None,
+    campaign_id: str | None,
+    selected_paths: Iterable[str] | None = None,
+    password: str | None = None,
+    conflict_strategy: str = "reject",
+    conflict_resolutions: Iterable[FileConflictResolution] | None = None,
+    metadata: dict[str, Any] | None = None,
+    is_admin: bool = False,
+    encryption_vault_id: str | None = None,
+    max_entries: int = ARCHIVE_UPLOAD_MAX_ENTRIES,
+    max_file_bytes: int = 50 * 1024 * 1024,
+    max_expanded_bytes: int = 2 * 1024 * 1024 * 1024,
+    max_expansion_ratio: int = 100,
+) -> list[UploadedStoredFile]:
+    inspection = inspect_archive(
+        archive_data,
+        filename=filename,
+        password=password,
+        max_entries=max_entries,
+        max_expanded_bytes=max_expanded_bytes,
+        max_expansion_ratio=max_expansion_ratio,
+    )
+    if inspection.requires_password and not inspection.password_verified:
+        raise ArchivePasswordError("Archive password is required")
+    selected_files = _selected_file_paths(inspection.entries, selected_paths)
+    if not selected_files:
+        raise FileStorageError("Select at least one archive file to import")
+    actual_total_limit = min(
+        max_expanded_bytes,
+        inspection.compressed_size_bytes * max_expansion_ratio,
+    )
+    if inspection.archive_format == "zip":
+        members = _read_selected_zip_members(
+            archive_data,
+            selected_files=selected_files,
+            password=password,
+            max_file_bytes=max_file_bytes,
+            max_total_bytes=actual_total_limit,
+        )
+    else:
+        members = _read_selected_tar_members(
+            archive_data,
+            selected_files=selected_files,
+            max_file_bytes=max_file_bytes,
+            max_total_bytes=actual_total_limit,
+        )
+    return _store_archive_members(
+        session,
+        members=members,
+        tenant_id=tenant_id,
+        owner_type=owner_type,
+        owner_id=owner_id,
+        user_id=user_id,
+        folder=folder,
+        campaign_id=campaign_id,
+        conflict_strategy=conflict_strategy,
+        conflict_resolutions=conflict_resolutions,
+        metadata=metadata,
+        is_admin=is_admin,
+        encryption_vault_id=encryption_vault_id,
+    )
+
+
 def extract_zip_upload(
     session: Session,
     *,
@@ -77,55 +254,409 @@ def extract_zip_upload(
     conflict_resolutions: Iterable[FileConflictResolution] | None = None,
     metadata: dict[str, Any] | None = None,
     is_admin: bool = False,
+    encryption_vault_id: str | None = None,
     max_files: int = ZIP_UPLOAD_MAX_FILES,
     max_file_bytes: int = 50 * 1024 * 1024,
     max_total_bytes: int = 250 * 1024 * 1024,
 ) -> list[UploadedStoredFile]:
-    uploaded: list[UploadedStoredFile] = []
+    """Backward-compatible wrapper for the original immediate ZIP endpoint."""
-    total = 0
+
-    base_folder = normalize_folder(folder)
+    return extract_archive_upload(
+        session,
+        tenant_id=tenant_id,
+        owner_type=owner_type,
+        owner_id=owner_id,
+        user_id=user_id,
+        archive_data=zip_data,
+        filename="archive.zip",
+        folder=folder,
+        campaign_id=campaign_id,
+        conflict_strategy=conflict_strategy,
+        conflict_resolutions=conflict_resolutions,
+        metadata=metadata,
+        is_admin=is_admin,
+        encryption_vault_id=encryption_vault_id,
+        max_entries=max_files,
+        max_file_bytes=max_file_bytes,
+        max_expanded_bytes=max_total_bytes,
+    )
+
+
+def _inspect_zip(
+    archive_data: bytes | str | PathLike[str],
+    *,
+    password: str | None,
+) -> tuple[list[ArchiveEntry], bool, bool]:
     try:
-        source = BytesIO(zip_data) if isinstance(zip_data, bytes) else zip_data
+        with pyzipper.AESZipFile(_archive_source(archive_data)) as archive:
-        with zipfile.ZipFile(source) as archive:
+            infos = archive.infolist()
-            infos = [info for info in archive.infolist() if not info.is_dir()]
+            entries = [_zip_entry(info) for info in infos]
-            if len(infos) > max_files:
+            encrypted_info = next(
-                raise FileStorageError(f"ZIP contains too many files (limit {max_files})")
+                (
-            for info in infos:
+                    info
-                if info.flag_bits & 0x1:
+                    for info in infos
-                    raise FileStorageError("Encrypted ZIP uploads are not supported")
+                    if not info.is_dir() and bool(info.flag_bits & 0x1)
-                if info.file_size < 0:
+                ),
-                    raise FileStorageError("Invalid ZIP member")
+                None,
-                if info.file_size > max_file_bytes:
+            )
-                    raise FileStorageError(f"ZIP member {info.filename!r} exceeds per-file limit")
+            if encrypted_info is None:
-                if total + info.file_size > max_total_bytes:
+                return entries, False, True
-                    raise FileStorageError("ZIP is too large after extraction")
+            if not password:
-                inner_path = normalize_logical_path(info.filename)
+                return entries, True, False
-                target_path = f"{base_folder}/{inner_path}" if base_folder else inner_path
+            try:
-                data, total = _read_zip_member(
+                with archive.open(
-                    archive,
+                    encrypted_info, pwd=password.encode("utf-8")
-                    info,
+                ) as member:
-                    max_file_bytes=max_file_bytes,
+                    member.read(1)
-                    max_total_bytes=max_total_bytes,
+            except (RuntimeError, ValueError, zipfile.BadZipFile) as exc:
-                    current_total=total,
+                raise ArchivePasswordError("Archive password is incorrect") from exc
-                )
+            return entries, True, True
-                uploaded.append(
+    except ArchivePasswordError:
-                    create_file_asset(
+        raise
-                        session,
+    except (OSError, ValueError, zipfile.BadZipFile) as exc:
-                        tenant_id=tenant_id,
+        raise FileStorageError("Invalid ZIP upload") from exc
-                        owner_type=owner_type,
+
-                        owner_id=owner_id,
+
-                        user_id=user_id,
+def _inspect_tar(
-                        filename=filename_from_path(inner_path),
+    archive_data: bytes | str | PathLike[str],
-                        data=data,
+) -> list[ArchiveEntry]:
-                        display_path=target_path,
+    try:
-                        content_type=mimetypes.guess_type(inner_path)[0] or "application/octet-stream",
+        with _open_tar(archive_data) as archive:
-                        metadata=metadata,
+            entries: list[ArchiveEntry] = []
-                        campaign_id=campaign_id,
+            for member in archive.getmembers():
-                        conflict_strategy=conflict_strategy,
+                path = _safe_member_path(member.name)
-                        conflict_resolutions=conflict_resolutions,
+                if member.isdir():
-                        is_admin=is_admin,
+                    entries.append(
+                        ArchiveEntry(path=path, kind="directory", size_bytes=0)
+                    )
+                    continue
+                if not member.isfile():
+                    raise FileStorageError(
+                        f"Archive member {member.name!r} is not a regular file or directory"
+                    )
+                entries.append(
+                    ArchiveEntry(
+                        path=path,
+                        kind="file",
+                        size_bytes=max(0, int(member.size)),
                     )
                 )
-    except zipfile.BadZipFile as exc:
+            return entries
-        raise FileStorageError("Invalid ZIP upload") from exc
+    except FileStorageError:
+        raise
+    except (OSError, tarfile.TarError) as exc:
+        raise FileStorageError("Invalid TAR upload") from exc
+
+
+def _zip_entry(info: zipfile.ZipInfo) -> ArchiveEntry:
+    path = _safe_member_path(info.filename)
+    unix_mode = (info.external_attr >> 16) & 0xFFFF
+    file_type = stat.S_IFMT(unix_mode)
+    if file_type and not (
+        stat.S_ISREG(unix_mode) or stat.S_ISDIR(unix_mode)
+    ):
+        raise FileStorageError(
+            f"Archive member {info.filename!r} is not a regular file or directory"
+        )
+    if info.file_size < 0 or info.compress_size < 0:
+        raise FileStorageError(f"Archive member {info.filename!r} has invalid size metadata")
+    return ArchiveEntry(
+        path=path,
+        kind="directory" if info.is_dir() else "file",
+        size_bytes=0 if info.is_dir() else int(info.file_size),
+        compressed_size_bytes=(
+            None if info.is_dir() else int(info.compress_size)
+        ),
+        encrypted=bool(info.flag_bits & 0x1),
+    )
+
+
+def _validate_archive_limits(
+    entries: list[ArchiveEntry],
+    *,
+    compressed_size: int,
+    max_entries: int,
+    max_expanded_bytes: int,
+    max_expansion_ratio: int,
+) -> None:
+    if len(entries) > max_entries:
+        raise FileStorageError(
+            f"Archive contains too many entries (limit {max_entries})"
+        )
+    seen: dict[str, str] = {}
+    expanded_size = 0
+    for entry in entries:
+        previous_kind = seen.get(entry.path)
+        if previous_kind is not None:
+            raise FileStorageError(
+                f"Archive contains duplicate path {entry.path!r}"
+            )
+        seen[entry.path] = entry.kind
+        if entry.kind == "file":
+            expanded_size += entry.size_bytes
+            if expanded_size > max_expanded_bytes:
+                raise FileStorageError(
+                    "Archive is too large after extraction "
+                    f"(limit {max_expanded_bytes} bytes)"
+                )
+    if expanded_size and (
+        compressed_size <= 0
+        or expanded_size > compressed_size * max_expansion_ratio
+    ):
+        raise FileStorageError(
+            "Archive expansion ratio exceeds "
+            f"{max_expansion_ratio}:1"
+        )
+
+
+def _with_derived_directories(
+    entries: list[ArchiveEntry],
+) -> list[ArchiveEntry]:
+    by_path = {entry.path: entry for entry in entries}
+    for entry in entries:
+        parent = PurePosixPath(entry.path).parent
+        while str(parent) not in {"", "."}:
+            path = str(parent)
+            existing = by_path.get(path)
+            if existing and existing.kind != "directory":
+                raise FileStorageError(
+                    f"Archive path {path!r} is both a file and a directory"
+                )
+            by_path.setdefault(
+                path,
+                ArchiveEntry(path=path, kind="directory", size_bytes=0),
+            )
+            parent = parent.parent
+    return sorted(
+        by_path.values(),
+        key=lambda entry: (
+            tuple(entry.path.casefold().split("/")),
+            entry.kind != "directory",
+        ),
+    )
+
+
+def _selected_file_paths(
+    entries: tuple[ArchiveEntry, ...],
+    selected_paths: Iterable[str] | None,
+) -> set[str]:
+    file_paths = {entry.path for entry in entries if entry.kind == "file"}
+    if selected_paths is None:
+        return file_paths
+    known = {entry.path: entry for entry in entries}
+    normalized = {_safe_member_path(path) for path in selected_paths}
+    unknown = normalized - known.keys()
+    if unknown:
+        raise FileStorageError(
+            f"Archive selection contains unknown path {sorted(unknown)[0]!r}"
+        )
+    selected_files: set[str] = set()
+    for path in normalized:
+        entry = known[path]
+        if entry.kind == "file":
+            selected_files.add(path)
+            continue
+        prefix = f"{path}/"
+        selected_files.update(
+            file_path
+            for file_path in file_paths
+            if file_path.startswith(prefix)
+        )
+    return selected_files
+
+
+def _read_selected_zip_members(
+    archive_data: bytes | str | PathLike[str],
+    *,
+    selected_files: set[str],
+    password: str | None,
+    max_file_bytes: int,
+    max_total_bytes: int,
+) -> list[tuple[str, bytes]]:
+    result: list[tuple[str, bytes]] = []
+    total = 0
+    try:
+        with pyzipper.AESZipFile(_archive_source(archive_data)) as archive:
+            for info in archive.infolist():
+                if info.is_dir():
+                    continue
+                path = _safe_member_path(info.filename)
+                if path not in selected_files:
+                    continue
+                pwd = password.encode("utf-8") if password else None
+                try:
+                    with archive.open(info, pwd=pwd) as source:
+                        data, total = _read_member(
+                            source,
+                            path=path,
+                            max_file_bytes=max_file_bytes,
+                            max_total_bytes=max_total_bytes,
+                            current_total=total,
+                        )
+                except (RuntimeError, ValueError, zipfile.BadZipFile) as exc:
+                    if info.flag_bits & 0x1:
+                        raise ArchivePasswordError(
+                            "Archive password is incorrect"
+                        ) from exc
+                    raise
+                result.append((path, data))
+    except (FileStorageError, ArchivePasswordError):
+        raise
+    except (OSError, ValueError, zipfile.BadZipFile) as exc:
+        raise FileStorageError("ZIP extraction failed") from exc
+    return result
+
+
+def _read_selected_tar_members(
+    archive_data: bytes | str | PathLike[str],
+    *,
+    selected_files: set[str],
+    max_file_bytes: int,
+    max_total_bytes: int,
+) -> list[tuple[str, bytes]]:
+    result: list[tuple[str, bytes]] = []
+    total = 0
+    try:
+        with _open_tar(archive_data) as archive:
+            for member in archive.getmembers():
+                if not member.isfile():
+                    continue
+                path = _safe_member_path(member.name)
+                if path not in selected_files:
+                    continue
+                source = archive.extractfile(member)
+                if source is None:
+                    raise FileStorageError(
+                        f"Archive member {path!r} could not be read"
+                    )
+                with source:
+                    data, total = _read_member(
+                        source,
+                        path=path,
+                        max_file_bytes=max_file_bytes,
+                        max_total_bytes=max_total_bytes,
+                        current_total=total,
+                    )
+                result.append((path, data))
+    except FileStorageError:
+        raise
+    except (OSError, tarfile.TarError) as exc:
+        raise FileStorageError("TAR extraction failed") from exc
+    return result
+
+
+def _read_member(
+    source: BinaryIO,
+    *,
+    path: str,
+    max_file_bytes: int,
+    max_total_bytes: int,
+    current_total: int,
+) -> tuple[bytes, int]:
+    parts: list[bytes] = []
+    actual_size = 0
+    while True:
+        read_size = min(
+            _ARCHIVE_READ_CHUNK_SIZE,
+            max_file_bytes + 1 - actual_size,
+        )
+        chunk = source.read(read_size)
+        if not chunk:
+            break
+        actual_size += len(chunk)
+        if actual_size > max_file_bytes:
+            raise FileStorageError(
+                f"Archive member {path!r} exceeds per-file limit"
+            )
+        if current_total + actual_size > max_total_bytes:
+            raise FileStorageError("Archive is too large after extraction")
+        parts.append(chunk)
+    return b"".join(parts), current_total + actual_size
+
+
+def _store_archive_members(
+    session: Session,
+    *,
+    members: Iterable[tuple[str, bytes]],
+    tenant_id: str,
+    owner_type: str,
+    owner_id: str,
+    user_id: str,
+    folder: str | None,
+    campaign_id: str | None,
+    conflict_strategy: str,
+    conflict_resolutions: Iterable[FileConflictResolution] | None,
+    metadata: dict[str, Any] | None,
+    is_admin: bool,
+    encryption_vault_id: str | None,
+) -> list[UploadedStoredFile]:
+    uploaded: list[UploadedStoredFile] = []
+    base_folder = normalize_folder(folder)
+    for inner_path, data in members:
+        target_path = (
+            f"{base_folder}/{inner_path}" if base_folder else inner_path
+        )
+        uploaded.append(
+            create_file_asset(
+                session,
+                tenant_id=tenant_id,
+                owner_type=owner_type,
+                owner_id=owner_id,
+                user_id=user_id,
+                filename=filename_from_path(inner_path),
+                data=data,
+                display_path=target_path,
+                content_type=mimetypes.guess_type(inner_path)[0]
+                or "application/octet-stream",
+                metadata=metadata,
+                campaign_id=campaign_id,
+                conflict_strategy=conflict_strategy,
+                conflict_resolutions=conflict_resolutions,
+                is_admin=is_admin,
+                encryption_vault_id=encryption_vault_id,
+            )
+        )
     return uploaded
+
+
+def _safe_member_path(value: str) -> str:
+    raw = str(value or "").replace("\\", "/").strip()
+    if (
+        not raw
+        or "\x00" in raw
+        or raw.startswith("/")
+        or _WINDOWS_DRIVE_RE.match(raw)
+    ):
+        raise FileStorageError(f"Unsafe archive member path {value!r}")
+    if any(part == ".." for part in raw.split("/")):
+        raise FileStorageError(f"Unsafe archive member path {value!r}")
+    try:
+        return normalize_logical_path(raw.rstrip("/"))
+    except ValueError as exc:
+        raise FileStorageError(f"Unsafe archive member path {value!r}") from exc
+
+
+def _archive_size(archive_data: bytes | str | PathLike[str]) -> int:
+    if isinstance(archive_data, bytes):
+        return len(archive_data)
+    try:
+        return Path(archive_data).stat().st_size
+    except OSError as exc:
+        raise FileStorageError("Archive upload could not be read") from exc
+
+
+def _archive_source(
+    archive_data: bytes | str | PathLike[str],
+) -> BytesIO | str | PathLike[str]:
+    return BytesIO(archive_data) if isinstance(archive_data, bytes) else archive_data
+
+
+def _open_tar(
+    archive_data: bytes | str | PathLike[str],
+) -> tarfile.TarFile:
+    if isinstance(archive_data, bytes):
+        return tarfile.open(fileobj=BytesIO(archive_data), mode="r:*")
+    try:
+        return tarfile.open(name=archive_data, mode="r:*")
+    except OSError as exc:
+        raise FileStorageError("Archive upload could not be read") from exc

src/govoplan_files/backend/storage/backends.py

@@ -1,207 +1,32 @@
 from __future__ import annotations
 
-from dataclasses import dataclass, field
+from govoplan_core.core.object_storage import (
-from pathlib import Path
+    LocalFilesystemStorageBackend,
-from typing import Iterable, Protocol
+    S3StorageBackend,
-
+    StorageBackend,
-from govoplan_core.security.outbound_http import (
+    StorageBackendError,
-    OutboundHttpError,
+    StorageObjectInfo,
-    response_limit,
+    StorageObjectMissing,
-    validate_unpinned_sdk_http_url,
+    StorageObjectPage,
+    configured_storage_backend,
 )
 
 from govoplan_files.backend.runtime import settings
 
 
-class StorageBackendError(RuntimeError):
-    pass
-
-
-class StorageBackend(Protocol):
-    name: str
-
-    def put_bytes(self, key: str, data: bytes, *, content_type: str | None = None) -> None: ...
-    def get_bytes(self, key: str) -> bytes: ...
-    def iter_bytes(self, key: str, *, chunk_size: int = 1024 * 1024) -> Iterable[bytes]: ...
-    def delete(self, key: str) -> None: ...
-    def exists(self, key: str) -> bool: ...
-
-
-@dataclass(slots=True)
-class LocalFilesystemStorageBackend:
-    root: Path
-    fallback_roots: tuple[Path, ...] = field(default_factory=tuple)
-    name: str = "local"
-
-    def __post_init__(self) -> None:
-        self.root = self.root.expanduser().resolve()
-        self.fallback_roots = tuple(root.expanduser().resolve() for root in self.fallback_roots if root)
-        self.root.mkdir(parents=True, exist_ok=True)
-
-    def _path_for_root(self, root: Path, key: str) -> Path:
-        path = (root / key).resolve()
-        if not path.is_relative_to(root):
-            raise StorageBackendError("Storage key escapes local storage root")
-        return path
-
-    def _path(self, key: str) -> Path:
-        return self._path_for_root(self.root, key)
-
-    def _readable_path(self, key: str) -> Path:
-        primary = self._path(key)
-        if primary.exists() and primary.is_file():
-            return primary
-        for root in self.fallback_roots:
-            candidate = self._path_for_root(root, key)
-            if candidate.exists() and candidate.is_file():
-                return candidate
-        raise StorageBackendError("Stored object does not exist")
-
-    def put_bytes(self, key: str, data: bytes, *, content_type: str | None = None) -> None:
-        path = self._path(key)
-        path.parent.mkdir(parents=True, exist_ok=True)
-        path.write_bytes(data)
-
-    def get_bytes(self, key: str) -> bytes:
-        return self._readable_path(key).read_bytes()
-
-    def iter_bytes(self, key: str, *, chunk_size: int = 1024 * 1024) -> Iterable[bytes]:
-        path = self._readable_path(key)
-        with path.open("rb") as handle:
-            while True:
-                chunk = handle.read(chunk_size)
-                if not chunk:
-                    break
-                yield chunk
-
-    def delete(self, key: str) -> None:
-        path = self._path(key)
-        if path.exists() and path.is_file():
-            path.unlink()
-
-    def exists(self, key: str) -> bool:
-        try:
-            self._readable_path(key)
-        except StorageBackendError:
-            return False
-        return True
-
-
-@dataclass(slots=True)
-class S3StorageBackend:
-    bucket: str
-    endpoint_url: str
-    region_name: str
-    access_key_id: str
-    secret_access_key: str
-    name: str = "s3"
-
-    @property
-    def client(self):
-        try:
-            endpoint_url = validate_unpinned_sdk_http_url(
-                self.endpoint_url,
-                label="File storage S3 endpoint",
-            )
-        except OutboundHttpError as exc:
-            raise StorageBackendError(str(exc)) from exc
-        try:
-            import boto3
-        except ModuleNotFoundError as exc:
-            raise StorageBackendError("boto3 is required for the S3 storage backend") from exc
-        return boto3.client(
-            "s3",
-            endpoint_url=endpoint_url,
-            region_name=self.region_name,
-            aws_access_key_id=self.access_key_id,
-            aws_secret_access_key=self.secret_access_key,
-        )
-
-    def put_bytes(self, key: str, data: bytes, *, content_type: str | None = None) -> None:
-        max_bytes = response_limit("file")
-        if len(data) > max_bytes:
-            raise StorageBackendError(f"Stored object exceeds the deployment limit of {max_bytes} bytes")
-        kwargs = {"Bucket": self.bucket, "Key": key, "Body": data}
-        if content_type:
-            kwargs["ContentType"] = content_type
-        self.client.put_object(**kwargs)
-
-    def get_bytes(self, key: str) -> bytes:
-        try:
-            obj = self.client.get_object(Bucket=self.bucket, Key=key)
-            max_bytes = response_limit("file")
-            body = obj["Body"]
-            try:
-                _reject_declared_object_size(obj, max_bytes=max_bytes)
-                data = body.read(max_bytes + 1)
-                if len(data) > max_bytes:
-                    raise StorageBackendError(f"Stored object exceeds the deployment limit of {max_bytes} bytes")
-                return data
-            finally:
-                if hasattr(body, "close"):
-                    body.close()
-        except Exception as exc:  # pragma: no cover - depends on S3 backend
-            raise StorageBackendError(str(exc)) from exc
-
-    def iter_bytes(self, key: str, *, chunk_size: int = 1024 * 1024) -> Iterable[bytes]:
-        try:
-            obj = self.client.get_object(Bucket=self.bucket, Key=key)
-            max_bytes = response_limit("file")
-            body = obj["Body"]
-            try:
-                _reject_declared_object_size(obj, max_bytes=max_bytes)
-                total = 0
-                while True:
-                    chunk = body.read(chunk_size)
-                    if not chunk:
-                        break
-                    total += len(chunk)
-                    if total > max_bytes:
-                        raise StorageBackendError(f"Stored object exceeds the deployment limit of {max_bytes} bytes")
-                    yield chunk
-            finally:
-                if hasattr(body, "close"):
-                    body.close()
-        except Exception as exc:  # pragma: no cover - depends on S3 backend
-            raise StorageBackendError(str(exc)) from exc
-
-    def delete(self, key: str) -> None:
-        self.client.delete_object(Bucket=self.bucket, Key=key)
-
-    def exists(self, key: str) -> bool:
-        try:
-            self.client.head_object(Bucket=self.bucket, Key=key)
-            return True
-        except Exception:
-            return False
-
-
-def _reject_declared_object_size(obj: object, *, max_bytes: int) -> None:
-    if not isinstance(obj, dict):
-        return
-    try:
-        declared_size = int(obj.get("ContentLength"))
-    except (TypeError, ValueError):
-        return
-    if declared_size > max_bytes:
-        raise StorageBackendError(f"Stored object exceeds the deployment limit of {max_bytes} bytes")
-
-
-def _fallback_roots() -> tuple[Path, ...]:
-    raw = getattr(settings, "file_storage_local_fallback_roots", "") or ""
-    return tuple(Path(item.strip()) for item in str(raw).split(",") if item.strip())
-
-
 def get_storage_backend() -> StorageBackend:
-    backend = settings.file_storage_backend.lower().strip()
+    """Return the deployment-owned backend for Files-managed objects."""
-    if backend in {"local", "filesystem", "fs"}:
+
-        return LocalFilesystemStorageBackend(Path(settings.file_storage_local_root), fallback_roots=_fallback_roots())
+    return configured_storage_backend(settings)
-    if backend in {"s3", "garage"}:
+
-        return S3StorageBackend(
+
-            bucket=settings.file_storage_s3_bucket or settings.s3_bucket,
+__all__ = [
-            endpoint_url=settings.file_storage_s3_endpoint_url or settings.s3_endpoint_url,
+    "LocalFilesystemStorageBackend",
-            region_name=settings.file_storage_s3_region or settings.s3_region,
+    "S3StorageBackend",
-            access_key_id=settings.file_storage_s3_access_key_id or settings.s3_access_key_id,
+    "StorageBackend",
-            secret_access_key=settings.file_storage_s3_secret_access_key or settings.s3_secret_access_key,
+    "StorageBackendError",
-        )
+    "StorageObjectInfo",
-    raise StorageBackendError(f"Unsupported file storage backend: {settings.file_storage_backend}")
+    "StorageObjectMissing",
+    "StorageObjectPage",
+    "get_storage_backend",
+]

src/govoplan_files/backend/storage/campaign_attachments.py

@@ -13,12 +13,16 @@ from typing import Any, Iterator
 from sqlalchemy.orm import Session
 
 from govoplan_files.backend.db.models import FileAsset, FileBlob, FileShare, FileVersion
-from govoplan_files.backend.storage.backends import StorageBackend, StorageBackendError, get_storage_backend
+from govoplan_files.backend.storage.backends import StorageBackend, get_storage_backend
-from govoplan_files.backend.storage.common import FileStorageError
 from govoplan_files.backend.storage.files import current_versions_and_blobs, get_asset_for_user, list_assets_for_user, share_files
 from govoplan_files.backend.storage.access import ensure_owner_access
 from govoplan_files.backend.storage.paths import normalize_folder, normalize_logical_path, safe_storage_component
 from govoplan_files.backend.storage.provenance import source_provenance_from_metadata, source_revision_from_metadata
+from govoplan_files.backend.storage.share_state import effective_file_share_clause
+from govoplan_files.backend.storage.integrity import (
+    ensure_blob_is_readable,
+    read_verified_blob_bytes,
+)
 
 
 MANAGED_SOURCE_PREFIX = "managed:"
@@ -165,7 +169,7 @@ def _campaign_linked_asset_ids(
                 FileShare.file_asset_id.in_(chunk),
                 FileShare.target_type == "campaign",
                 FileShare.target_id == campaign_id,
-                FileShare.revoked_at.is_(None),
+                effective_file_share_clause(),
             )
             .all()
         )
@@ -301,11 +305,9 @@ def _materialize_managed_asset(
     target = _safe_local_target(local_root, relative_path)
     target.parent.mkdir(parents=True, exist_ok=True)
     version, blob = version_blobs[asset.id]
+    ensure_blob_is_readable(blob)
     if include_bytes:
-        try:
+        data = read_verified_blob_bytes(blob, backend=backend) if backend else b""
-            data = backend.get_bytes(blob.storage_key) if backend else b""
-        except StorageBackendError as exc:
-            raise FileStorageError(str(exc)) from exc
         target.write_bytes(data)
     else:
         target.touch()

src/govoplan_files/backend/storage/connector_browse.py

@@ -15,8 +15,8 @@ from defusedxml import ElementTree as SafeElementTree
 
 from govoplan_core.security.outbound_http import (
     OutboundHttpError,
-    validate_unpinned_sdk_host,
+    validate_outbound_host,
-    validate_unpinned_sdk_http_url,
+    validate_outbound_http_url,
 )
 
 from govoplan_files.backend.storage.http_client import ConnectorHttpError, request_connector_bytes
@@ -27,6 +27,12 @@ from govoplan_files.backend.storage.connector_deployment import (
     validate_connector_tls_metadata,
 )
 from govoplan_files.backend.storage.connector_profiles import ConnectorProfile
+from govoplan_files.backend.storage.sdk_peer_pinning import (
+    SdkPeerPinningError,
+    create_pinned_s3_client,
+    install_pinned_smb_transport,
+    pinned_smb_connection_cache,
+)
 
 
 class ConnectorBrowseError(RuntimeError):
@@ -246,6 +252,28 @@ def _browse_smb(profile: ConnectorProfile, *, path: str) -> list[ConnectorBrowse
 
 def _browse_s3(profile: ConnectorProfile, *, path: str, library_id: str | None, continuation_token: str | None) -> list[ConnectorBrowseItem]:
     client = _s3_client(profile)
+    try:
+        return _browse_s3_with_client(
+            client,
+            profile=profile,
+            path=path,
+            library_id=library_id,
+            continuation_token=continuation_token,
+        )
+    finally:
+        close = getattr(client, "close", None)
+        if callable(close):
+            close()
+
+
+def _browse_s3_with_client(
+    client: Any,
+    *,
+    profile: ConnectorProfile,
+    path: str,
+    library_id: str | None,
+    continuation_token: str | None,
+) -> list[ConnectorBrowseItem]:
     bucket = _s3_bucket(profile, library_id)
     if not bucket:
         try:
@@ -313,23 +341,22 @@ def _s3_client(profile: ConnectorProfile) -> Any:
         raise ConnectorBrowseError("Secret-ref S3 credentials need a runtime secret resolver before live browsing")
     if profile.endpoint_url:
         try:
-            endpoint_url = validate_unpinned_sdk_http_url(
+            endpoint_url = validate_outbound_http_url(
                 profile.endpoint_url,
                 label="S3 connector endpoint",
             )
         except OutboundHttpError as exc:
             raise ConnectorBrowseError(str(exc)) from exc
     else:
-        raise ConnectorBrowseError(
+        endpoint_url = None
-            "S3 connector endpoint discovery uses an SDK transport that cannot guarantee connection-time DNS/IP "
-            "pinning; live S3 access is disabled until that transport supports pinning"
-        )
     try:
-        boto3 = import_module("boto3")
         config_module = import_module("botocore.config")
+        unsigned = import_module("botocore").UNSIGNED
     except ImportError as exc:
         raise ConnectorBrowseUnsupported("S3 connector browsing requires the optional boto3 dependency") from exc
-    kwargs: dict[str, object] = {"endpoint_url": endpoint_url}
+    kwargs: dict[str, object] = {}
+    if endpoint_url:
+        kwargs["endpoint_url"] = endpoint_url
     region = _metadata_string(profile, "region") or _metadata_string(profile, "aws_region")
     if region:
         kwargs["region_name"] = region
@@ -346,10 +373,21 @@ def _s3_client(profile: ConnectorProfile) -> Any:
     if verify is not None:
         kwargs["verify"] = verify
     addressing_style = _s3_addressing_style(profile)
+    config_values: dict[str, object] = {
+        "proxies": {},
+        "retries": {"mode": "standard", "max_attempts": 4},
+    }
     if addressing_style:
-        kwargs["config"] = config_module.Config(s3={"addressing_style": addressing_style})
+        config_values["s3"] = {"addressing_style": addressing_style}
+    if bool(access_key) != bool(secret_key):
+        raise ConnectorBrowseError("S3 connectors require both an access key and a secret key")
+    if not access_key:
+        config_values["signature_version"] = unsigned
+    kwargs["config"] = config_module.Config(**config_values)
     try:
-        return boto3.client("s3", **kwargs)
+        return create_pinned_s3_client(**kwargs)
+    except SdkPeerPinningError as exc:
+        raise ConnectorBrowseError(str(exc)) from exc
     except Exception as exc:  # pragma: no cover - concrete exception types are dependency-version specific
         raise ConnectorBrowseError(f"S3 connector could not be initialized: {exc}") from exc
 
@@ -783,7 +821,7 @@ def _smb_location(profile: ConnectorProfile) -> _SmbLocation:
         raise ConnectorBrowseError("SMB connector endpoint_url must include a server")
     port = parsed.port or _int(profile.metadata.get("port")) or 445
     try:
-        validate_unpinned_sdk_host(server, port=port, label="SMB connector endpoint")
+        validate_outbound_host(server, port=port, label="SMB connector endpoint")
     except OutboundHttpError as exc:
         raise ConnectorBrowseError(str(exc)) from exc
     path_parts = [part for part in unquote(parsed.path or "").strip("/").split("/") if part]
@@ -810,6 +848,7 @@ def _smb_unc_path(location: _SmbLocation, path: str) -> str:
 def _smb_client_kwargs(profile: ConnectorProfile, location: _SmbLocation) -> dict[str, object]:
     kwargs: dict[str, object] = {
         "port": location.port,
+        "connection_cache": pinned_smb_connection_cache(),
         "require_signing": _metadata_bool(profile, "require_signing", default=True),
         "auth_protocol": _metadata_string(profile, "auth_protocol") or "ntlm",
     }
@@ -845,9 +884,11 @@ def _profile_token(profile: ConnectorProfile) -> str | None:
 
 def _smbclient_module() -> Any:
     try:
-        return import_module("smbclient")
+        return install_pinned_smb_transport(import_module("smbclient"))
     except ImportError as exc:
         raise ConnectorBrowseUnsupported("SMB connector browsing requires the optional smbprotocol dependency") from exc
+    except SdkPeerPinningError as exc:
+        raise ConnectorBrowseUnsupported(str(exc)) from exc
 
 
 def _smb_entry_stat(entry: object) -> object | None:

src/govoplan_files/backend/storage/connector_credential_store.py

@@ -5,8 +5,17 @@ from dataclasses import dataclass
 from typing import Any
 
 from sqlalchemy.orm import Session
+from sqlalchemy import inspect
 
 from govoplan_core.core.policy import normalize_policy_scope_type, policy_source_path
+from govoplan_core.security.credential_envelopes import (
+    CredentialAccessContext,
+    CredentialEnvelope,
+    CredentialEnvelopeError,
+    ResolvedCredentialEnvelope,
+    list_credential_envelopes,
+    resolve_credential_envelope,
+)
 from govoplan_core.security.secrets import decrypt_secret, encrypt_secret
 from govoplan_files.backend.db.models import FileConnectorCredential
 from govoplan_files.backend.storage.common import FileStorageError
@@ -14,6 +23,8 @@ from govoplan_files.backend.storage.connector_deployment import reject_api_contr
 from govoplan_files.backend.storage.connector_policy import ConnectorPolicySource, connector_policy_sources_from_payload
 from govoplan_files.backend.storage.connector_profiles import supported_connector_providers
 
+CORE_CREDENTIAL_ENVELOPE_PREFIX = "credential-envelope:"
+
 
 @dataclass(frozen=True, slots=True)
 class ConnectorCredential:
@@ -33,6 +44,7 @@ class ConnectorCredential:
     policy_sources: tuple[ConnectorPolicySource, ...] = ()
     metadata: Mapping[str, Any] | None = None
     source_kind: str = "database"
+    has_secret: bool = False
 
     @property
     def source_path(self) -> str:
@@ -58,7 +70,7 @@ class ConnectorCredential:
             return True
         # Environment references are deliberately unavailable to API-managed
         # credential rows. Legacy rows remain visible but fail closed.
-        return bool(self.secret_ref or self.password_value or self.token_value)
+        return bool(self.has_secret or self.secret_ref or self.password_value or self.token_value)
 
     def to_response(self) -> dict[str, Any]:
         return {
@@ -85,7 +97,159 @@ def list_database_connector_credentials(
     tenant_id: str,
     include_disabled: bool = False,
 ) -> list[ConnectorCredential]:
-    return [connector_credential_from_row(row) for row in list_connector_credential_rows(session, tenant_id=tenant_id, include_disabled=include_disabled)]
+    local = [
+        connector_credential_from_row(row)
+        for row in list_connector_credential_rows(
+            session,
+            tenant_id=tenant_id,
+            include_disabled=include_disabled,
+        )
+    ]
+    return [
+        *local,
+        *list_reusable_connector_credentials(
+            session,
+            tenant_id=tenant_id,
+            include_disabled=include_disabled,
+        ),
+    ]
+
+
+def reusable_credential_reference(credential_id: str) -> str:
+    return f"{CORE_CREDENTIAL_ENVELOPE_PREFIX}{credential_id}"
+
+
+def reusable_credential_id(credential_ref: str | None) -> str | None:
+    if not credential_ref or not credential_ref.startswith(CORE_CREDENTIAL_ENVELOPE_PREFIX):
+        return None
+    value = credential_ref.removeprefix(CORE_CREDENTIAL_ENVELOPE_PREFIX).strip()
+    return value or None
+
+
+def file_credential_context(
+    *,
+    tenant_id: str,
+    profile_id: str | None = None,
+    scope_type: str = "tenant",
+    scope_id: str | None = None,
+    administrative: bool = False,
+) -> CredentialAccessContext:
+    return CredentialAccessContext(
+        tenant_id=tenant_id,
+        user_id=scope_id if scope_type == "user" else None,
+        group_ids=frozenset({scope_id}) if scope_type == "group" and scope_id else frozenset(),
+        target_scope_type=scope_type,
+        target_scope_id=scope_id or tenant_id,
+        module_id="files",
+        server_ref=f"files:{profile_id}" if profile_id else None,
+        administrative=administrative,
+    )
+
+
+def list_reusable_connector_credentials(
+    session: Session,
+    *,
+    tenant_id: str,
+    include_disabled: bool = False,
+) -> list[ConnectorCredential]:
+    if not inspect(session.get_bind()).has_table(CredentialEnvelope.__tablename__):
+        return []
+    context = file_credential_context(tenant_id=tenant_id, administrative=True)
+    return [
+        connector_credential_from_envelope(row)
+        for row in list_credential_envelopes(
+            session,
+            context=context,
+            include_inactive=include_disabled,
+        )
+    ]
+
+
+def resolve_reusable_connector_credential(
+    session: Session,
+    *,
+    tenant_id: str,
+    credential_ref: str,
+    profile_id: str,
+    scope_type: str,
+    scope_id: str | None,
+) -> ConnectorCredential:
+    credential_id = reusable_credential_id(credential_ref)
+    if credential_id is None:
+        raise FileStorageError("Reusable credential reference is invalid")
+    try:
+        resolved = resolve_credential_envelope(
+            session,
+            credential_id=credential_id,
+            context=file_credential_context(
+                tenant_id=tenant_id,
+                profile_id=profile_id,
+                scope_type=scope_type,
+                scope_id=scope_id,
+            ),
+        )
+    except CredentialEnvelopeError as exc:
+        raise FileStorageError("Reusable credential is unavailable to this file connection") from exc
+    return connector_credential_from_resolved_envelope(
+        resolved,
+        scope_type=scope_type,
+        scope_id=scope_id,
+    )
+
+
+def connector_credential_from_envelope(row: CredentialEnvelope) -> ConnectorCredential:
+    return ConnectorCredential(
+        id=reusable_credential_reference(row.id),
+        label=row.name,
+        scope_type=row.scope_type,
+        scope_id=row.scope_id,
+        enabled=row.is_active,
+        credential_mode=_envelope_credential_mode(row.credential_kind, row.secret_keys),
+        username=_clean_public_value(row.public_data, "username"),
+        source_kind="credential_envelope",
+        has_secret=bool(row.secret_data_encrypted),
+        metadata={
+            "credential_envelope_id": row.id,
+            "credential_kind": row.credential_kind,
+            "allowed_modules": list(row.allowed_modules or []),
+            "allowed_server_refs": list(row.allowed_server_refs or []),
+            "inherit_to_lower_scopes": bool(row.inherit_to_lower_scopes),
+            "revision": row.revision,
+        },
+    )
+
+
+def connector_credential_from_resolved_envelope(
+    row: ResolvedCredentialEnvelope,
+    *,
+    scope_type: str,
+    scope_id: str | None,
+) -> ConnectorCredential:
+    return ConnectorCredential(
+        id=reusable_credential_reference(row.id),
+        label=row.name,
+        scope_type=scope_type,
+        scope_id=scope_id,
+        enabled=True,
+        credential_mode=_envelope_credential_mode(row.credential_kind, row.secret_data),
+        username=_clean_public_value(row.public_data, "username"),
+        password_value=_first_secret(row.secret_data, "password", "secret"),
+        token_value=_first_secret(
+            row.secret_data,
+            "access_token",
+            "bearer_token",
+            "token",
+            "api_key",
+        ),
+        source_kind="credential_envelope",
+        has_secret=bool(row.secret_data),
+        metadata={
+            "credential_envelope_id": row.id,
+            "credential_kind": row.credential_kind,
+            "inherit_to_lower_scopes": True,
+            "revision": row.revision,
+        },
+    )
 
 
 def list_connector_credential_rows(
@@ -348,6 +512,27 @@ def _policy_source_response(source: ConnectorPolicySource) -> dict[str, Any]:
     }
 
 
+def _envelope_credential_mode(kind: str, secret_values: Mapping[str, Any] | list[str]) -> str:
+    keys = {str(key) for key in secret_values}
+    if kind in {"token", "oauth2", "api_key"} or keys.intersection(
+        {"access_token", "bearer_token", "token", "api_key"}
+    ):
+        return "token"
+    return "basic"
+
+
+def _clean_public_value(values: Mapping[str, Any] | None, key: str) -> str | None:
+    return _clean((values or {}).get(key))
+
+
+def _first_secret(values: Mapping[str, Any], *keys: str) -> str | None:
+    for key in keys:
+        value = _clean(values.get(key))
+        if value is not None:
+            return value
+    return None
+
+
 def _clean(value: object) -> str | None:
     if value is None:
         return None

src/govoplan_files/backend/storage/connector_folder_sync.py

@@ -0,0 +1,139 @@
+from __future__ import annotations
+
+from collections import deque
+from dataclasses import dataclass, field
+
+from govoplan_files.backend.storage.connector_browse import (
+    ConnectorBrowseItem,
+    browse_connector_profile,
+    normalize_connector_browse_path,
+)
+from govoplan_files.backend.storage.connector_profiles import ConnectorProfile
+
+
+@dataclass(frozen=True, slots=True)
+class ConnectorFolderSkip:
+    path: str
+    reason: str
+
+
+@dataclass(slots=True)
+class ConnectorFolderDiscovery:
+    files: list[ConnectorBrowseItem] = field(default_factory=list)
+    skipped: list[ConnectorFolderSkip] = field(default_factory=list)
+    truncated: bool = False
+
+
+def discover_connector_folder(
+    profile: ConnectorProfile,
+    *,
+    path: str,
+    library_id: str | None,
+    recursive: bool,
+    max_files: int,
+    max_depth: int,
+) -> ConnectorFolderDiscovery:
+    """Discover a bounded, deterministic set of files below one connector folder."""
+
+    root_path = normalize_connector_browse_path(path)
+    queue: deque[tuple[str, int]] = deque([(root_path, 0)])
+    visited_folders: set[str] = set()
+    seen_files: set[tuple[str, str]] = set()
+    result = ConnectorFolderDiscovery()
+
+    while queue:
+        folder_path, depth = queue.popleft()
+        if folder_path in visited_folders:
+            continue
+        visited_folders.add(folder_path)
+
+        continuation_token: str | None = None
+        seen_tokens: set[str] = set()
+        while True:
+            items = browse_connector_profile(
+                profile,
+                path=folder_path,
+                library_id=library_id,
+                continuation_token=continuation_token,
+            )
+            for item in items:
+                item_path = normalize_connector_browse_path(item.path)
+                if item.kind == "file":
+                    identity = (str(item.external_id or ""), item_path)
+                    if identity in seen_files:
+                        continue
+                    if len(result.files) >= max_files:
+                        result.truncated = True
+                        result.skipped.append(
+                            ConnectorFolderSkip(
+                                path=item_path or folder_path,
+                                reason=f"The manual sync limit of {max_files} files was reached.",
+                            )
+                        )
+                        return result
+                    seen_files.add(identity)
+                    result.files.append(item)
+                    continue
+
+                if item.kind == "folder":
+                    if not recursive:
+                        result.skipped.append(
+                            ConnectorFolderSkip(
+                                path=item_path,
+                                reason="Subfolder skipped because recursive sync is disabled.",
+                            )
+                        )
+                    elif depth >= max_depth:
+                        result.skipped.append(
+                            ConnectorFolderSkip(
+                                path=item_path,
+                                reason=f"Subfolder skipped at the configured depth limit of {max_depth}.",
+                            )
+                        )
+                    else:
+                        queue.append((item_path, depth + 1))
+                    continue
+
+                result.skipped.append(
+                    ConnectorFolderSkip(
+                        path=item_path,
+                        reason=f"Connector item kind {item.kind!r} cannot be synchronized as a file.",
+                    )
+                )
+
+            next_token = _next_continuation_token(items)
+            if not next_token or next_token in seen_tokens:
+                break
+            seen_tokens.add(next_token)
+            continuation_token = next_token
+
+    result.files.sort(key=lambda item: normalize_connector_browse_path(item.path).casefold())
+    result.skipped.sort(key=lambda item: (item.path.casefold(), item.reason))
+    return result
+
+
+def connector_relative_path(*, space_root: str, item_path: str) -> str:
+    root = normalize_connector_browse_path(space_root)
+    item = normalize_connector_browse_path(item_path)
+    if not root:
+        return item
+    if item == root:
+        return item.rsplit("/", 1)[-1]
+    prefix = f"{root}/"
+    if not item.startswith(prefix):
+        raise ValueError("Connector returned a file outside the linked connector-space root")
+    return item[len(prefix) :]
+
+
+def join_connector_path(*parts: str | None) -> str:
+    return normalize_connector_browse_path(
+        "/".join(str(part or "").strip("/\\") for part in parts if str(part or "").strip("/\\"))
+    )
+
+
+def _next_continuation_token(items: list[ConnectorBrowseItem]) -> str | None:
+    for item in reversed(items):
+        value = item.metadata.get("next_continuation_token")
+        if value is not None and str(value).strip():
+            return str(value).strip()
+    return None

src/govoplan_files/backend/storage/connector_imports.py

@@ -314,15 +314,20 @@ def _read_s3_file(profile: ConnectorProfile, *, library_id: str, path: str, max_
     if not key:
         raise ConnectorImportError("S3 import requires an object key")
     client = _s3_import_client(profile)
-    detail = _s3_object_detail(client, bucket=bucket, key=key, max_bytes=max_bytes)
+    try:
-    version_id = _clean(detail.get("VersionId"))
+        detail = _s3_object_detail(client, bucket=bucket, key=key, max_bytes=max_bytes)
-    response, data = _download_s3_object(
+        version_id = _clean(detail.get("VersionId"))
-        client,
+        response, data = _download_s3_object(
-        bucket=bucket,
+            client,
-        key=key,
+            bucket=bucket,
-        version_id=version_id,
+            key=key,
-        max_bytes=max_bytes,
+            version_id=version_id,
-    )
+            max_bytes=max_bytes,
+        )
+    finally:
+        close = getattr(client, "close", None)
+        if callable(close):
+            close()
     content_type = _clean(response.get("ContentType") if isinstance(response, dict) else None) or _clean(detail.get("ContentType")) or mimetypes.guess_type(key)[0]
     etag = _clean(response.get("ETag") if isinstance(response, dict) else None) or _clean(detail.get("ETag"))
     filename = filename_from_path(key)

src/govoplan_files/backend/storage/connector_policy.py

@@ -246,11 +246,11 @@ def _applied_fields(policy: Mapping[str, Any]) -> tuple[str, ...]:
 
 def _matches_field(request: ConnectorAccessRequest, field: str, patterns: list[str]) -> bool:
     if field == "connectors":
-        return _matches_exact(request.connector_id, patterns)
+        return _matches_reference(request.connector_id, patterns)
     if field == "credentials":
-        return _matches_exact(request.credential_id, patterns)
+        return _matches_reference(request.credential_id, patterns)
     if field == "providers":
-        return _matches_exact(request.provider, patterns)
+        return _matches_reference(request.provider, patterns)
     if field == "external_ids":
         return _matches_glob(request.external_id, patterns)
     if field == "external_paths":
@@ -260,11 +260,11 @@ def _matches_field(request: ConnectorAccessRequest, field: str, patterns: list[s
     return False
 
 
-def _matches_exact(value: str | None, patterns: list[str]) -> bool:
+def _matches_reference(value: str | None, patterns: list[str]) -> bool:
     if value is None:
         return False
     clean = value.casefold()
-    return any(pattern == "*" or clean == pattern.casefold() for pattern in patterns)
+    return any(fnmatchcase(clean, pattern.casefold()) for pattern in patterns)
 
 
 def _matches_glob(value: str | None, patterns: list[str]) -> bool:

src/govoplan_files/backend/storage/connector_profile_store.py

@@ -10,7 +10,13 @@ from govoplan_core.security.secrets import decrypt_secret, encrypt_secret
 from govoplan_files.backend.db.models import FileConnectorCredential, FileConnectorProfile
 from govoplan_files.backend.storage.common import FileStorageError
 from govoplan_files.backend.storage.connector_deployment import reject_api_controlled_deployment_references
-from govoplan_files.backend.storage.connector_credential_store import connector_credential_from_row, credential_rows_by_id
+from govoplan_files.backend.storage.connector_credential_store import (
+    ConnectorCredential,
+    connector_credential_from_row,
+    credential_rows_by_id,
+    resolve_reusable_connector_credential,
+    reusable_credential_id,
+)
 from govoplan_files.backend.storage.connector_policy import connector_policy_sources_from_payload
 from govoplan_files.backend.storage.connector_profiles import ConnectorProfile, supported_connector_providers
 
@@ -50,10 +56,28 @@ def select_database_connector_profiles(
     profile_ids = {row.id for row in rows}
     if row_visible is not None:
         rows = [row for row in rows if row_visible(row)]
-    credential_ids = {_clean(row.credential_profile_id) for row in rows if _clean(row.credential_profile_id)}
+    credential_ids = {
+        _clean(row.credential_profile_id)
+        for row in rows
+        if _clean(row.credential_profile_id) and not reusable_credential_id(row.credential_profile_id)
+    }
     credentials = credential_rows_by_id(session, tenant_id=tenant_id, credential_ids={item for item in credential_ids if item}, include_disabled=include_disabled)
     return (
-        [connector_profile_from_row(row, credential_row=credentials.get(row.credential_profile_id or "")) for row in rows],
+        [
+            connector_profile_from_row(
+                row,
+                credential_row=(
+                    _resolve_profile_reusable_credential(
+                        session,
+                        tenant_id=tenant_id,
+                        row=row,
+                    )
+                    if reusable_credential_id(row.credential_profile_id)
+                    else credentials.get(row.credential_profile_id or "")
+                ),
+            )
+            for row in rows
+        ],
         profile_ids,
     )
 
@@ -73,7 +97,10 @@ def list_connector_profile_rows(
     return query.order_by(FileConnectorProfile.scope_type.asc(), FileConnectorProfile.label.asc()).all()
 
 
-def connector_profile_from_row(row: FileConnectorProfile, credential_row: FileConnectorCredential | None = None) -> ConnectorProfile:
+def connector_profile_from_row(
+    row: FileConnectorProfile,
+    credential_row: FileConnectorCredential | ConnectorCredential | None = None,
+) -> ConnectorProfile:
     policy_sources = []
     if row.policy:
         policy_sources = connector_policy_sources_from_payload({
@@ -82,9 +109,18 @@ def connector_profile_from_row(row: FileConnectorProfile, credential_row: FileCo
             "label": row.label,
             "policy": row.policy,
         })
-    credential = connector_credential_from_row(credential_row) if credential_row is not None else None
+    credential = (
+        credential_row
+        if isinstance(credential_row, ConnectorCredential)
+        else connector_credential_from_row(credential_row)
+        if credential_row is not None
+        else None
+    )
     if credential:
         policy_sources.extend(credential.policy_sources)
+    metadata = dict(row.metadata_ or {})
+    if reusable_credential_id(row.credential_profile_id) and credential is None:
+        metadata["credential_unavailable"] = True
     return ConnectorProfile(
         id=row.id,
         label=row.label,
@@ -105,11 +141,30 @@ def connector_profile_from_row(row: FileConnectorProfile, credential_row: FileCo
         token_value=credential.token_value if credential else decrypt_secret(row.token_encrypted),
         capabilities=tuple(_string_list(row.capabilities)),
         policy_sources=tuple(policy_sources),
-        metadata=dict(row.metadata_ or {}),
+        metadata=metadata,
         source_kind="database",
     )
 
 
+def _resolve_profile_reusable_credential(
+    session: Session,
+    *,
+    tenant_id: str,
+    row: FileConnectorProfile,
+) -> ConnectorCredential | None:
+    try:
+        return resolve_reusable_connector_credential(
+            session,
+            tenant_id=tenant_id,
+            credential_ref=row.credential_profile_id or "",
+            profile_id=row.id,
+            scope_type=row.scope_type,
+            scope_id=row.scope_id,
+        )
+    except FileStorageError:
+        return None
+
+
 def get_connector_profile_row(
     session: Session,
     *,

src/govoplan_files/backend/storage/connector_providers.py

@@ -32,6 +32,7 @@ class ConnectorProviderDescriptor:
             "installed": self.installed,
             "browse_supported": self.browse_supported,
             "import_supported": self.import_supported,
+            "write_supported": self.provider == "s3",
             "optional_dependency": self.optional_dependency,
             "permission_model": self.permission_model,
             "sync_strategy": self.sync_strategy,
@@ -110,7 +111,7 @@ def connector_provider_descriptors() -> tuple[ConnectorProviderDescriptor, ...]:
             conflict_strategy="Managed import/sync uses existing files conflict_strategy handling after download.",
             preview_strategy="Previews are generated from the frozen managed file after import or sync, not directly from the share.",
             audit_events=("files.connector.imported", "files.connector.synced", "files.connector.accessed"),
-            notes="Browse/import logic is implemented, but live smbprotocol access fails closed until the SDK supports connection-time DNS/IP pinning for initial connections and DFS referrals.",
+            notes="Initial sessions, reconnects, aliases, and DFS referral targets use the Files-owned pinned smbprotocol transport and the deployment-wide private-network policy.",
         ),
         ConnectorProviderDescriptor(
             provider="s3",
@@ -125,7 +126,7 @@ def connector_provider_descriptors() -> tuple[ConnectorProviderDescriptor, ...]:
             conflict_strategy="Managed import/sync uses existing files conflict_strategy handling after object download.",
             preview_strategy="Previews are generated from the frozen managed file after import or sync, not directly from the bucket.",
             audit_events=("files.connector.imported", "files.connector.synced", "files.connector.accessed"),
-            notes="Browse/import logic is implemented, but live boto3 access fails closed until its HTTP transport supports connection-time DNS/IP pinning and redirect revalidation.",
+            notes="Every botocore connection pool uses the Files-owned pinned transport, including retries, redirects, endpoint discovery, and virtual-host aliases; outbound proxies and ambient credential discovery are disabled.",
         ),
         ConnectorProviderDescriptor(
             provider="sharepoint",

src/govoplan_files/backend/storage/connector_spaces.py

@@ -14,6 +14,7 @@ from govoplan_files.backend.storage.connector_profiles import ConnectorProfile
 
 
 SYNC_MODES = {"manual"}
+WRITABLE_PROVIDERS = {"s3"}
 
 
 def connector_space_owner_id(space: FileConnectorSpace) -> str:
@@ -40,6 +41,7 @@ def create_connector_space(
     ensure_owner_access(session, tenant_id=tenant_id, owner_type=owner_type, owner_id=owner_id, user_id=user_id, is_admin=is_admin)
     label = _normalize_label(label)
     sync_mode = _normalize_sync_mode(sync_mode)
+    validate_connector_space_write_mode(profile, read_only=read_only)
     remote_path = normalize_connector_browse_path(remote_path)
     library_id = _clean_optional(library_id)
 
@@ -145,6 +147,7 @@ def update_connector_space(
     library_id: str | None = None,
     remote_path: str | None = None,
     sync_mode: str | None = None,
+    read_only: bool | None = None,
     is_active: bool | None = None,
     metadata: Mapping[str, Any] | None = None,
     is_admin: bool = False,
@@ -178,6 +181,8 @@ def update_connector_space(
         space.remote_path = normalize_connector_browse_path(remote_path)
     if sync_mode is not None:
         space.sync_mode = _normalize_sync_mode(sync_mode)
+    if read_only is not None:
+        space.read_only = bool(read_only)
     if is_active is not None:
         space.is_active = bool(is_active)
     if metadata is not None:
@@ -187,6 +192,17 @@ def update_connector_space(
     return space
 
 
+def validate_connector_space_write_mode(
+    profile: ConnectorProfile, *, read_only: bool
+) -> None:
+    if read_only:
+        return
+    if profile.provider not in WRITABLE_PROVIDERS or "write" not in profile.capabilities:
+        raise FileStorageError(
+            "Two-way mode requires an S3 connection with the write capability enabled"
+        )
+
+
 def soft_delete_connector_space(
     session: Session,
     space: FileConnectorSpace,

src/govoplan_files/backend/storage/connector_visibility.py

@@ -136,11 +136,6 @@ def connector_profile_usable_for_import(profile: ConnectorProfile) -> bool:
         and descriptor.import_supported
     ):
         return False
-    # These SDK paths intentionally fail closed until every connection peer and
-    # SDK-managed redirect/referral can be pinned and revalidated.
-    if profile.provider in {"s3", "smb"}:
-        return False
-
     # Prove that the initial root browse performed by the current Files UI is
     # policy-allowed. A later selected remote path/item is checked again.
     return connector_policy_decision(

src/govoplan_files/backend/storage/connector_writes.py

@@ -0,0 +1,438 @@
+from __future__ import annotations
+
+from dataclasses import dataclass
+import hashlib
+from typing import Any, Mapping
+
+from govoplan_core.core.recovery import (
+    RecoveryGuaranteeError,
+    RecoveryMode,
+    RecoveryPlan,
+    RecoveryStatus,
+)
+from govoplan_core.core.recovery_runtime import (
+    RecoveryOperationBusy,
+    RecoveryOperationStateConflict,
+    begin_durable_recovery_operation,
+)
+from govoplan_core.core.runtime_coordination import process_runtime_identity
+from govoplan_core.db.session import get_database
+from govoplan_files.backend.storage.common import FileStorageError
+from govoplan_files.backend.storage.connector_browse import (
+    ConnectorBrowseError,
+    ConnectorBrowseUnsupported,
+    _clean,
+    _s3_bucket,
+    _s3_client,
+    _s3_object_key,
+    normalize_connector_browse_path,
+)
+from govoplan_files.backend.storage.connector_profiles import ConnectorProfile
+
+
+@dataclass(frozen=True, slots=True)
+class ConnectorWriteResult:
+    recovery_operation_id: str
+    status: str
+    replayed: bool
+    provider: str
+    remote_path: str
+    revision: str | None
+    checksum_sha256: str
+    size_bytes: int
+
+
+@dataclass(frozen=True, slots=True)
+class _HeadProbe:
+    observed: Mapping[str, Any] | None
+    verified: bool
+    exception_type: str | None = None
+
+
+def write_connector_file(
+    profile: ConnectorProfile,
+    *,
+    tenant_id: str,
+    library_id: str | None,
+    remote_path: str,
+    data: bytes,
+    content_type: str | None,
+    expected_revision: str | None,
+    idempotency_key: str,
+) -> ConnectorWriteResult:
+    if profile.provider != "s3" or "write" not in profile.capabilities:
+        raise FileStorageError(
+            "Remote writes require an S3 connection with the write capability enabled"
+        )
+    try:
+        normalized_path = normalize_connector_browse_path(remote_path)
+    except ConnectorBrowseError as exc:
+        raise FileStorageError(str(exc)) from exc
+    if not normalized_path:
+        raise FileStorageError("Remote write requires an object path")
+    try:
+        bucket = _s3_bucket(profile, library_id)
+        key = _s3_object_key(profile, normalized_path)
+    except ConnectorBrowseError as exc:
+        raise FileStorageError(str(exc)) from exc
+    if not bucket or not key:
+        raise FileStorageError("S3 remote write requires a bucket and object path")
+    checksum = hashlib.sha256(data).hexdigest()
+    target_digest = hashlib.sha256(f"{bucket}:{key}".encode("utf-8")).hexdigest()
+    request = {
+        "tenant_id": tenant_id,
+        "connector_profile_id": profile.id,
+        "provider": profile.provider,
+        "remote_target_sha256": target_digest,
+        "content_sha256": checksum,
+        "size_bytes": len(data),
+        "expected_revision": expected_revision,
+    }
+    try:
+        started = begin_durable_recovery_operation(
+            get_database().SessionLocal,
+            identity=process_runtime_identity(),
+            module_id="files",
+            operation_type="connector-s3-write",
+            idempotency_key=f"files-connector-write:{idempotency_key}",
+            request=request,
+            recovery_plan=RecoveryPlan(
+                mode=RecoveryMode.FORWARD_RECOVERY,
+                preconditions=(
+                    "the connector space is explicitly configured for two-way writes",
+                    "the connection and inherited policy allow the exact remote path",
+                    "a conditional create or expected remote revision prevents blind overwrite",
+                ),
+                forward_recovery_steps=(
+                    "inspect provider metadata for the request and content digests",
+                    "resolve success only for an exact digest match or retry after verified absence",
+                ),
+                verification_steps=(
+                    "read S3 object metadata after the write",
+                    "match the recorded request and content digests without downloading content",
+                ),
+            ),
+            precondition_evidence={
+                "remote_target_sha256": target_digest,
+                "content_sha256": checksum,
+                "size_bytes": len(data),
+                "conditional_write": True,
+            },
+            lease_resource_key=f"files:connector:{tenant_id}:{profile.id}:{target_digest[:40]}",
+            lease_ttl_seconds=15 * 60,
+            resource_type="file_connector_object",
+            resource_id=target_digest,
+            metadata={
+                "resources": ["postgresql", "s3-connector"],
+                "provider": "s3",
+                "connector_profile_id": profile.id,
+            },
+            block_unresolved_resource=True,
+        )
+    except (RecoveryOperationBusy, RecoveryOperationStateConflict) as exc:
+        raise FileStorageError(
+            "This remote object is owned by another write or unresolved recovery operation"
+        ) from exc
+    except (RecoveryGuaranteeError, RuntimeError, ValueError) as exc:
+        raise FileStorageError(
+            "The Files recovery ledger is unavailable; the remote object was not changed"
+        ) from exc
+    if started.replayed or started.operation is None:
+        return ConnectorWriteResult(
+            recovery_operation_id=started.operation_id,
+            status=started.status,
+            replayed=True,
+            provider="s3",
+            remote_path=normalized_path,
+            revision=None,
+            checksum_sha256=checksum,
+            size_bytes=len(data),
+        )
+    operation = started.operation
+    try:
+        client = _s3_client(profile)
+    except (ConnectorBrowseError, ConnectorBrowseUnsupported) as exc:
+        operation.reject(
+            summary="The S3 client was unavailable before the remote effect",
+            evidence=_verified(
+                {"external_effect_started": False}, exception_type=type(exc).__name__
+            ),
+        )
+        raise FileStorageError(str(exc)) from exc
+    try:
+        try:
+            before = _head_object(client, bucket=bucket, key=key)
+        except Exception as exc:
+            operation.reject(
+                summary="Remote preconditions could not be inspected before the effect",
+                evidence=_verified(
+                    {"external_effect_started": False},
+                    exception_type=type(exc).__name__,
+                ),
+            )
+            raise FileStorageError("S3 remote revision lookup failed") from exc
+        expected = _normalize_revision(expected_revision)
+        if before is not None and expected is None:
+            operation.reject(
+                summary="A blind remote overwrite was rejected",
+                evidence=_verified(
+                    {"remote_object_exists": True, "expected_revision_supplied": False}
+                ),
+            )
+            raise FileStorageError(
+                "The remote object already exists; reload it and supply its expected revision"
+            )
+        if expected is not None and (
+            before is None or expected not in _observed_revisions(before)
+        ):
+            operation.reject(
+                summary="The expected remote revision did not match",
+                evidence=_verified(
+                    {
+                        "remote_object_exists": before is not None,
+                        "expected_revision_matches": False,
+                    }
+                ),
+            )
+            raise FileStorageError("The remote object changed; reload before writing")
+        params: dict[str, object] = {
+            "Bucket": bucket,
+            "Key": key,
+            "Body": data,
+            "Metadata": {
+                "govoplan-sha256": checksum,
+                "govoplan-request-id": started.operation_id,
+            },
+        }
+        if content_type:
+            params["ContentType"] = content_type
+        if before is None:
+            params["IfNoneMatch"] = "*"
+        else:
+            params["IfMatch"] = str(before.get("ETag") or expected_revision or "")
+        try:
+            client.put_object(**params)
+        except Exception as exc:
+            probe = _probe_head_object(client, bucket=bucket, key=key)
+            if probe.verified and _matches_effect(
+                probe.observed,
+                checksum=checksum,
+                operation_id=started.operation_id,
+            ):
+                operation.succeed(
+                    evidence=_success_evidence(
+                        probe.observed,
+                        checksum=checksum,
+                        operation_id=started.operation_id,
+                    )
+                )
+                return _result(
+                    started.operation_id,
+                    normalized_path,
+                    checksum,
+                    len(data),
+                    probe.observed,
+                )
+            operation.unresolved(
+                status=RecoveryStatus.OUTCOME_UNKNOWN,
+                summary="The S3 write outcome requires reconciliation",
+                evidence={
+                    "remote_target_sha256": target_digest,
+                    "observed": _public_observation(probe),
+                    "exception_type": type(exc).__name__,
+                },
+                failure_summary="Inspect the remote object metadata before retrying this path",
+            )
+            return ConnectorWriteResult(
+                recovery_operation_id=started.operation_id,
+                status=RecoveryStatus.OUTCOME_UNKNOWN.value,
+                replayed=False,
+                provider="s3",
+                remote_path=normalized_path,
+                revision=_revision(probe.observed),
+                checksum_sha256=checksum,
+                size_bytes=len(data),
+            )
+        probe = _probe_head_object(client, bucket=bucket, key=key)
+        if not probe.verified:
+            operation.unresolved(
+                status=RecoveryStatus.OUTCOME_UNKNOWN,
+                summary="The S3 write returned but provider evidence could not be queried",
+                evidence={
+                    "remote_target_sha256": target_digest,
+                    "observed": _public_observation(probe),
+                },
+                failure_summary="Inspect the remote object metadata before retrying this path",
+            )
+            return ConnectorWriteResult(
+                recovery_operation_id=started.operation_id,
+                status=RecoveryStatus.OUTCOME_UNKNOWN.value,
+                replayed=False,
+                provider="s3",
+                remote_path=normalized_path,
+                revision=None,
+                checksum_sha256=checksum,
+                size_bytes=len(data),
+            )
+        observed = probe.observed
+        if not _matches_effect(
+            observed, checksum=checksum, operation_id=started.operation_id
+        ):
+            operation.unresolved(
+                status=RecoveryStatus.RECOVERY_REQUIRED,
+                summary="The S3 write returned but exact provider evidence did not match",
+                evidence={
+                    "remote_target_sha256": target_digest,
+                    "observed": _public_observation(probe),
+                },
+                failure_summary="Reconcile the remote object before another write",
+            )
+            return ConnectorWriteResult(
+                recovery_operation_id=started.operation_id,
+                status=RecoveryStatus.RECOVERY_REQUIRED.value,
+                replayed=False,
+                provider="s3",
+                remote_path=normalized_path,
+                revision=_revision(observed),
+                checksum_sha256=checksum,
+                size_bytes=len(data),
+            )
+        operation.succeed(
+            evidence=_success_evidence(
+                observed, checksum=checksum, operation_id=started.operation_id
+            )
+        )
+        return _result(
+            started.operation_id,
+            normalized_path,
+            checksum,
+            len(data),
+            observed,
+        )
+    finally:
+        close = getattr(client, "close", None)
+        if callable(close):
+            close()
+
+
+def _head_object(client: Any, *, bucket: str, key: str) -> Mapping[str, Any] | None:
+    try:
+        response = client.head_object(Bucket=bucket, Key=key)
+    except Exception as exc:
+        if _is_not_found(exc):
+            return None
+        raise
+    if not isinstance(response, Mapping):
+        raise FileStorageError("S3 connector returned invalid object metadata")
+    return response
+
+
+def _probe_head_object(client: Any, *, bucket: str, key: str) -> _HeadProbe:
+    try:
+        return _HeadProbe(
+            observed=_head_object(client, bucket=bucket, key=key), verified=True
+        )
+    except Exception as exc:
+        return _HeadProbe(
+            observed=None,
+            verified=False,
+            exception_type=type(exc).__name__,
+        )
+
+
+def _is_not_found(exc: Exception) -> bool:
+    response = getattr(exc, "response", None)
+    if not isinstance(response, Mapping):
+        return False
+    error = response.get("Error")
+    code = error.get("Code") if isinstance(error, Mapping) else None
+    status = response.get("ResponseMetadata")
+    http_status = status.get("HTTPStatusCode") if isinstance(status, Mapping) else None
+    return str(code).casefold() in {"404", "nosuchkey", "notfound"} or http_status == 404
+
+
+def _metadata(observed: Mapping[str, Any] | None) -> Mapping[str, Any]:
+    value = observed.get("Metadata") if observed else None
+    return value if isinstance(value, Mapping) else {}
+
+
+def _matches_effect(
+    observed: Mapping[str, Any] | None, *, checksum: str, operation_id: str
+) -> bool:
+    metadata = _metadata(observed)
+    return bool(
+        observed is not None
+        and _clean(metadata.get("govoplan-sha256")) == checksum
+        and _clean(metadata.get("govoplan-request-id")) == operation_id
+    )
+
+
+def _revision(observed: Mapping[str, Any] | None) -> str | None:
+    if observed is None:
+        return None
+    return _clean(observed.get("VersionId") or observed.get("ETag"))
+
+
+def _normalize_revision(value: str | None) -> str | None:
+    cleaned = _clean(value)
+    return cleaned.strip('"') if cleaned else None
+
+
+def _observed_revisions(observed: Mapping[str, Any]) -> set[str]:
+    return {
+        normalized
+        for value in (observed.get("VersionId"), observed.get("ETag"))
+        if (normalized := _normalize_revision(_clean(value))) is not None
+    }
+
+
+def _success_evidence(
+    observed: Mapping[str, Any] | None, *, checksum: str, operation_id: str
+) -> dict[str, object]:
+    return _verified(
+        {
+            "remote_object_present": observed is not None,
+            "content_digest_matches": _clean(_metadata(observed).get("govoplan-sha256")) == checksum,
+            "request_marker_matches": _clean(_metadata(observed).get("govoplan-request-id")) == operation_id,
+        },
+        revision=_revision(observed),
+        content_sha256=checksum,
+    )
+
+
+def _public_observation(probe: _HeadProbe) -> dict[str, object]:
+    observed = probe.observed
+    return {
+        "probe_verified": probe.verified,
+        "probe_exception_type": probe.exception_type,
+        "present": observed is not None,
+        "revision": _revision(observed),
+        "has_content_digest": bool(_clean(_metadata(observed).get("govoplan-sha256"))),
+        "has_request_marker": bool(_clean(_metadata(observed).get("govoplan-request-id"))),
+    }
+
+
+def _verified(checks: dict[str, object], **details: object) -> dict[str, object]:
+    return {"verified": True, "checks": checks, **details}
+
+
+def _result(
+    operation_id: str,
+    remote_path: str,
+    checksum: str,
+    size_bytes: int,
+    observed: Mapping[str, Any] | None,
+) -> ConnectorWriteResult:
+    return ConnectorWriteResult(
+        recovery_operation_id=operation_id,
+        status=RecoveryStatus.SUCCEEDED.value,
+        replayed=False,
+        provider="s3",
+        remote_path=remote_path,
+        revision=_revision(observed),
+        checksum_sha256=checksum,
+        size_bytes=size_bytes,
+    )
+
+
+__all__ = ["ConnectorWriteResult", "write_connector_file"]

src/govoplan_files/backend/storage/content_protection.py

@@ -0,0 +1,93 @@
+from __future__ import annotations
+
+from govoplan_core.core.encryption import (
+    ContentProtectionRequest,
+    ContentUnprotectionRequest,
+    ProtectedContent,
+    encryption_content_cipher,
+)
+from govoplan_files.backend.runtime import get_registry
+from govoplan_files.backend.storage.common import FileStorageError
+
+
+FILES_PROTECTION_PROFILE = "files-server-envelope-v1"
+
+
+def protect_blob_content(
+    session: object,
+    *,
+    tenant_id: str,
+    blob_id: str,
+    vault_id: str,
+    ciphertext_ref: str,
+    plaintext: bytes,
+    actor_id: str,
+    content_type: str | None,
+) -> ProtectedContent:
+    capability = encryption_content_cipher(get_registry())
+    if capability is None:
+        raise FileStorageError(
+            "File encryption was requested, but the Encryption module is unavailable."
+        )
+    try:
+        return capability.protect_content(
+            session,
+            request=ContentProtectionRequest(
+                tenant_id=tenant_id,
+                owner_module="files",
+                resource_type="file_blob",
+                resource_id=blob_id,
+                profile_id=FILES_PROTECTION_PROFILE,
+                vault_id=vault_id,
+                ciphertext_ref=ciphertext_ref,
+                plaintext=plaintext,
+                policy_decision_ref="files:explicit-vault-selection:v1",
+                idempotency_key=f"file-blob:{blob_id}:content:v1",
+                actor_id=actor_id,
+                metadata={
+                    "content_type": content_type or "application/octet-stream",
+                },
+            ),
+        )
+    except Exception as exc:
+        raise FileStorageError(
+            "Managed file content could not be protected by the configured vault."
+        ) from exc
+
+
+def unprotect_blob_content(
+    session: object,
+    *,
+    tenant_id: str,
+    blob_id: str,
+    envelope_id: str,
+    ciphertext: bytes,
+) -> bytes:
+    capability = encryption_content_cipher(get_registry())
+    if capability is None:
+        raise FileStorageError(
+            "This file is encrypted and cannot be read while Encryption is unavailable."
+        )
+    try:
+        return capability.unprotect_content(
+            session,
+            request=ContentUnprotectionRequest(
+                tenant_id=tenant_id,
+                owner_module="files",
+                resource_type="file_blob",
+                resource_id=blob_id,
+                envelope_id=envelope_id,
+                ciphertext=ciphertext,
+            ),
+        )
+    except Exception as exc:
+        raise FileStorageError(
+            "Managed file content could not be opened with its protection envelope."
+        ) from exc
+
+
+__all__ = [
+    "FILES_PROTECTION_PROFILE",
+    "protect_blob_content",
+    "unprotect_blob_content",
+]

src/govoplan_files/backend/storage/files.py

@@ -2,21 +2,32 @@ from __future__ import annotations
 
 import hashlib
 import mimetypes
+from datetime import datetime
 from pathlib import PurePosixPath
 from typing import Any, Iterable
 from uuid import uuid4
 
-from sqlalchemy import and_, or_
+from sqlalchemy import and_, exists, func, or_
 from sqlalchemy.orm import Session
 
 from govoplan_core.core.campaigns import CAPABILITY_CAMPAIGNS_ACCESS, CampaignAccessProvider
 from govoplan_files.backend.db.models import CampaignAttachmentUse, FileAsset, FileBlob, FileShare, FileVersion
 from govoplan_files.backend.runtime import get_registry, settings
 from govoplan_files.backend.storage.access import ensure_owner_access, ensure_share_target_exists, user_group_ids
-from govoplan_files.backend.storage.backends import StorageBackendError, get_storage_backend
+from govoplan_files.backend.storage.backends import (
+    StorageBackendError,
+    StorageObjectMissing,
+    get_storage_backend,
+)
 from govoplan_files.backend.storage.common import FileConflictResolution, FileStorageError, UploadedStoredFile, utcnow
-from govoplan_files.backend.storage.paths import filename_from_path, join_folder_filename, normalize_folder, normalize_logical_path, safe_storage_component
+from govoplan_files.backend.storage.paths import filename_from_path, join_folder_filename, normalize_folder, normalize_logical_path
 from govoplan_files.backend.storage.provenance import source_provenance_from_metadata
+from govoplan_files.backend.storage.recovery import begin_blob_write_recovery
+from govoplan_files.backend.storage.integrity import (
+    QUARANTINED_BLOB_STATUSES,
+    read_verified_blob_bytes,
+)
+from govoplan_files.backend.storage.share_state import effective_file_share_clause
 
 
 def _campaign_access_provider() -> CampaignAccessProvider:
@@ -51,8 +62,9 @@ def _storage_backend_name() -> str:
     return settings.file_storage_backend.lower().strip()
 
 
-def _storage_key(*, tenant_id: str, checksum: str, filename: str) -> str:
+def _storage_key(*, tenant_id: str, checksum: str) -> str:
-    return f"tenants/{tenant_id}/files/{checksum[:2]}/{uuid4().hex}-{safe_storage_component(filename)}"
+    # Object locators remain opaque so recovery evidence never persists names.
+    return f"tenants/{tenant_id}/files/{checksum[:2]}/{uuid4().hex}.blob"
 
 
 def _get_or_create_blob(
@@ -62,40 +74,144 @@ def _get_or_create_blob(
     data: bytes,
     filename: str,
     content_type: str | None,
+    actor_id: str,
+    encryption_vault_id: str | None = None,
 ) -> FileBlob:
     checksum = hashlib.sha256(data).hexdigest()
     size = len(data)
+    vault_id = str(encryption_vault_id or "").strip() or None
+    protection_discriminator = f"vault:{vault_id}" if vault_id else "plaintext"
     blob = (
         session.query(FileBlob)
-        .filter(FileBlob.tenant_id == tenant_id, FileBlob.checksum_sha256 == checksum, FileBlob.size_bytes == size)
+        .filter(FileBlob.tenant_id == tenant_id, FileBlob.checksum_sha256 == checksum, FileBlob.size_bytes == size, FileBlob.protection_discriminator == protection_discriminator)
+        .with_for_update()
         .one_or_none()
     )
     if blob:
         backend = get_storage_backend()
-        if not backend.exists(blob.storage_key):
+        repair_required = (
+            blob.integrity_status in QUARANTINED_BLOB_STATUSES
+            or blob.quarantined_at is not None
+        )
+        try:
+            backend.stat(blob.storage_key)
+        except StorageObjectMissing:
+            repair_required = True
+        except StorageBackendError as exc:
+            raise FileStorageError(str(exc)) from exc
+        if repair_required:
+            repair_token = hashlib.sha256(
+                repr(
+                    (
+                        blob.integrity_status,
+                        blob.integrity_checked_at,
+                        blob.quarantined_at,
+                        blob.storage_checksum_sha256,
+                    )
+                ).encode("utf-8")
+            ).hexdigest()
+            recovery = begin_blob_write_recovery(
+                session,
+                backend=backend,
+                tenant_id=tenant_id,
+                blob_id=blob.id,
+                storage_key=blob.storage_key,
+                semantic_checksum_sha256=checksum,
+                semantic_size_bytes=size,
+                protection_discriminator=protection_discriminator,
+                created_new=False,
+                repair_token=repair_token,
+            )
+            stored_data = data
+            expected_envelope_id = blob.encryption_envelope_id
+            if vault_id:
+                from govoplan_files.backend.storage.content_protection import protect_blob_content
+
+                protected = protect_blob_content(
+                    session,
+                    tenant_id=tenant_id,
+                    blob_id=blob.id,
+                    vault_id=vault_id,
+                    ciphertext_ref=blob.storage_key,
+                    plaintext=data,
+                    actor_id=actor_id,
+                    content_type=content_type,
+                )
+                if blob.encryption_envelope_id not in {None, protected.envelope.envelope_id}:
+                    raise FileStorageError("The existing encrypted blob has another protection envelope.")
+                stored_data = protected.ciphertext
+                blob.encryption_envelope_id = protected.envelope.envelope_id
+                expected_envelope_id = protected.envelope.envelope_id
+                blob.storage_checksum_sha256 = hashlib.sha256(stored_data).hexdigest()
+                blob.storage_size_bytes = len(stored_data)
+            recovery.prepare_stored_bytes(
+                stored_data,
+                envelope_id=expected_envelope_id,
+            )
             try:
-                backend.put_bytes(blob.storage_key, data, content_type=content_type)
+                backend.put_bytes(blob.storage_key, stored_data, content_type="application/octet-stream" if vault_id else content_type)
             except StorageBackendError as exc:
                 raise FileStorageError(str(exc)) from exc
+            blob.integrity_status = "verified"
+            blob.integrity_checked_at = utcnow()
+            blob.integrity_failure = None
+            blob.quarantined_at = None
         blob.ref_count += 1
         session.add(blob)
         return blob
 
-    storage_key = _storage_key(tenant_id=tenant_id, checksum=checksum, filename=filename)
+    blob_id = str(uuid4())
+    storage_key = _storage_key(tenant_id=tenant_id, checksum=checksum)
     backend = get_storage_backend()
+    recovery = begin_blob_write_recovery(
+        session,
+        backend=backend,
+        tenant_id=tenant_id,
+        blob_id=blob_id,
+        storage_key=storage_key,
+        semantic_checksum_sha256=checksum,
+        semantic_size_bytes=size,
+        protection_discriminator=protection_discriminator,
+        created_new=True,
+    )
+    stored_data = data
+    envelope_id = None
+    if vault_id:
+        from govoplan_files.backend.storage.content_protection import protect_blob_content
+
+        protected = protect_blob_content(
+            session,
+            tenant_id=tenant_id,
+            blob_id=blob_id,
+            vault_id=vault_id,
+            ciphertext_ref=storage_key,
+            plaintext=data,
+            actor_id=actor_id,
+            content_type=content_type,
+        )
+        stored_data = protected.ciphertext
+        envelope_id = protected.envelope.envelope_id
+    recovery.prepare_stored_bytes(stored_data, envelope_id=envelope_id)
     try:
-        backend.put_bytes(storage_key, data, content_type=content_type)
+        backend.put_bytes(storage_key, stored_data, content_type="application/octet-stream" if vault_id else content_type)
     except StorageBackendError as exc:
         raise FileStorageError(str(exc)) from exc
     blob = FileBlob(
+        id=blob_id,
         tenant_id=tenant_id,
         storage_backend=_storage_backend_name(),
         storage_bucket=_storage_bucket_name(),
         storage_key=storage_key,
         checksum_sha256=checksum,
         size_bytes=size,
+        protection_discriminator=protection_discriminator,
+        encryption_envelope_id=envelope_id,
+        storage_checksum_sha256=hashlib.sha256(stored_data).hexdigest() if vault_id else None,
+        storage_size_bytes=len(stored_data) if vault_id else None,
         content_type=content_type,
         ref_count=1,
+        integrity_status="verified",
+        integrity_checked_at=utcnow(),
     )
     session.add(blob)
     session.flush()
@@ -120,6 +236,7 @@ def create_file_asset(
     conflict_strategy: str = "reject",
     conflict_resolutions: Iterable[FileConflictResolution] | None = None,
     is_admin: bool = False,
+    encryption_vault_id: str | None = None,
 ) -> UploadedStoredFile:
     owner_type = owner_type.lower().strip()
     ensure_owner_access(session, tenant_id=tenant_id, owner_type=owner_type, owner_id=owner_id, user_id=user_id, is_admin=is_admin)
@@ -147,7 +264,7 @@ def create_file_asset(
     elif action == "rename":
         logical_path = _next_available_logical_path(session, tenant_id=tenant_id, owner_type=owner_type, owner_id=owner_id, desired_path=logical_path)
 
-    blob = _get_or_create_blob(session, tenant_id=tenant_id, data=data, filename=safe_filename, content_type=content_type)
+    blob = _get_or_create_blob(session, tenant_id=tenant_id, data=data, filename=safe_filename, content_type=content_type, actor_id=user_id, encryption_vault_id=encryption_vault_id)
     asset = FileAsset(
         tenant_id=tenant_id,
         owner_type=owner_type,
@@ -279,6 +396,7 @@ def update_file_asset_content(
     data: bytes,
     content_type: str | None,
     metadata: dict[str, Any],
+    encryption_vault_id: str | None = None,
 ) -> tuple[UploadedStoredFile, str]:
     if asset.tenant_id != tenant_id or asset.deleted_at is not None:
         raise FileStorageError("File not found")
@@ -289,10 +407,23 @@ def update_file_asset_content(
     checksum = hashlib.sha256(data).hexdigest()
     asset.metadata_ = metadata
     session.add(asset)
-    if current_blob.checksum_sha256 == checksum and current_blob.size_bytes == len(data):
+    inherited_vault_id = encryption_vault_id
+    if inherited_vault_id is None and current_blob.encryption_envelope_id:
+        prefix = "vault:"
+        if current_blob.protection_discriminator.startswith(prefix):
+            inherited_vault_id = current_blob.protection_discriminator[len(prefix) :]
+    inherited_vault_id = str(inherited_vault_id or "").strip() or None
+    target_protection = (
+        f"vault:{inherited_vault_id}" if inherited_vault_id else "plaintext"
+    )
+    if (
+        current_blob.checksum_sha256 == checksum
+        and current_blob.size_bytes == len(data)
+        and current_blob.protection_discriminator == target_protection
+    ):
         return UploadedStoredFile(asset=asset, version=current_version, blob=current_blob), "unchanged"
 
-    blob = _get_or_create_blob(session, tenant_id=tenant_id, data=data, filename=safe_filename, content_type=content_type)
+    blob = _get_or_create_blob(session, tenant_id=tenant_id, data=data, filename=safe_filename, content_type=content_type, actor_id=user_id, encryption_vault_id=inherited_vault_id)
     version = FileVersion(
         tenant_id=tenant_id,
         file_asset_id=asset.id,
@@ -328,7 +459,7 @@ def get_asset_for_user(session: Session, *, tenant_id: str, user_id: str, asset_
         .filter(
             FileShare.tenant_id == tenant_id,
             FileShare.file_asset_id == asset.id,
-            FileShare.revoked_at.is_(None),
+            effective_file_share_clause(),
             FileShare.permission.in_(permission_values),
             or_(
                 (FileShare.target_type == "user") & (FileShare.target_id == user_id),
@@ -343,6 +474,32 @@ def get_asset_for_user(session: Session, *, tenant_id: str, user_id: str, asset_
     return asset
 
 
+def get_asset_for_share_management(
+    session: Session,
+    *,
+    tenant_id: str,
+    user_id: str,
+    asset_id: str,
+    is_admin: bool = False,
+) -> FileAsset:
+    asset = session.get(FileAsset, asset_id)
+    if not asset or asset.tenant_id != tenant_id or asset.deleted_at is not None:
+        raise FileStorageError("File not found")
+    if is_admin:
+        return asset
+    owns_asset = (
+        asset.owner_type == "user"
+        and asset.owner_user_id == user_id
+    ) or (
+        asset.owner_type == "group"
+        and asset.owner_group_id
+        in user_group_ids(session, tenant_id=tenant_id, user_id=user_id)
+    )
+    if not owns_asset:
+        raise FileStorageError("Only file owners and administrators can manage shares")
+    return asset
+
+
 def list_assets_for_user(
     session: Session,
     *,
@@ -352,6 +509,8 @@ def list_assets_for_user(
     owner_id: str | None = None,
     campaign_id: str | None = None,
     path_prefix: str | None = None,
+    campaign_usage: str | None = None,
+    audit_relevant: bool | None = None,
     include_deleted: bool = False,
     is_admin: bool = False,
 ) -> list[FileAsset]:
@@ -363,12 +522,49 @@ def list_assets_for_user(
         owner_id=owner_id,
         campaign_id=campaign_id,
         path_prefix=path_prefix,
+        campaign_usage=campaign_usage,
+        audit_relevant=audit_relevant,
         include_deleted=include_deleted,
         is_admin=is_admin,
     )
     return query.order_by(FileAsset.display_path.asc(), FileAsset.updated_at.desc(), FileAsset.id.asc()).all()
 
 
+def list_recent_assets_for_user(
+    session: Session,
+    *,
+    tenant_id: str,
+    user_id: str,
+    limit: int,
+    owner_type: str | None = None,
+    owner_id: str | None = None,
+    campaign_id: str | None = None,
+    path_prefix: str | None = None,
+    campaign_usage: str | None = None,
+    audit_relevant: bool | None = None,
+    is_admin: bool = False,
+) -> list[FileAsset]:
+    """Return a bounded recent projection through the normal access query."""
+
+    query = _asset_visibility_query_for_user(
+        session,
+        tenant_id=tenant_id,
+        user_id=user_id,
+        owner_type=owner_type,
+        owner_id=owner_id,
+        campaign_id=campaign_id,
+        path_prefix=path_prefix,
+        campaign_usage=campaign_usage,
+        audit_relevant=audit_relevant,
+        is_admin=is_admin,
+    )
+    return (
+        query.order_by(FileAsset.updated_at.desc(), FileAsset.id.asc())
+        .limit(max(1, limit))
+        .all()
+    )
+
+
 def list_assets_for_user_window(
     session: Session,
     *,
@@ -378,6 +574,8 @@ def list_assets_for_user_window(
     owner_id: str | None = None,
     campaign_id: str | None = None,
     path_prefix: str | None = None,
+    campaign_usage: str | None = None,
+    audit_relevant: bool | None = None,
     include_deleted: bool = False,
     is_admin: bool = False,
     page_size: int,
@@ -393,6 +591,8 @@ def list_assets_for_user_window(
         owner_id=owner_id,
         campaign_id=campaign_id,
         path_prefix=path_prefix,
+        campaign_usage=campaign_usage,
+        audit_relevant=audit_relevant,
         include_deleted=include_deleted,
         is_admin=is_admin,
     )
@@ -417,6 +617,8 @@ def _asset_visibility_query_for_user(
     owner_id: str | None = None,
     campaign_id: str | None = None,
     path_prefix: str | None = None,
+    campaign_usage: str | None = None,
+    audit_relevant: bool | None = None,
     include_deleted: bool = False,
     is_admin: bool = False,
 ):
@@ -430,30 +632,97 @@ def _asset_visibility_query_for_user(
         if owner_type == "group" and owner_id:
             query = query.filter(FileAsset.owner_group_id == owner_id)
     if campaign_id:
-        query = query.join(FileShare, FileShare.file_asset_id == FileAsset.id).filter(
+        campaign_share = exists().where(
             FileShare.tenant_id == tenant_id,
+            FileShare.file_asset_id == FileAsset.id,
             FileShare.target_type == "campaign",
             FileShare.target_id == campaign_id,
-            FileShare.revoked_at.is_(None),
+            effective_file_share_clause(),
         )
+        query = query.filter(campaign_share)
     elif not is_admin and not owner_type:
         group_ids = user_group_ids(session, tenant_id=tenant_id, user_id=user_id)
-        query = query.outerjoin(FileShare, FileShare.file_asset_id == FileAsset.id).filter(
+        active_share = exists().where(
+            FileShare.tenant_id == tenant_id,
+            FileShare.file_asset_id == FileAsset.id,
+            effective_file_share_clause(),
+            or_(
+                (FileShare.target_type == "user") & (FileShare.target_id == user_id),
+                (FileShare.target_type == "group") & (FileShare.target_id.in_(group_ids)),
+                (FileShare.target_type == "tenant") & (FileShare.target_id == tenant_id),
+            )
+        )
+        query = query.filter(
             or_(
                 (FileAsset.owner_type == "user") & (FileAsset.owner_user_id == user_id),
                 (FileAsset.owner_type == "group") & (FileAsset.owner_group_id.in_(group_ids)),
-                (FileShare.revoked_at.is_(None)) & (FileShare.target_type == "user") & (FileShare.target_id == user_id),
+                active_share,
-                (FileShare.revoked_at.is_(None)) & (FileShare.target_type == "group") & (FileShare.target_id.in_(group_ids)),
-                (FileShare.revoked_at.is_(None)) & (FileShare.target_type == "tenant") & (FileShare.target_id == tenant_id),
             )
         )
     if path_prefix:
         prefix = normalize_folder(path_prefix)
         if prefix:
             query = query.filter(FileAsset.display_path.like(f"{prefix}/%"))
+    campaign_share_exists = exists().where(
+        FileShare.tenant_id == tenant_id,
+        FileShare.file_asset_id == FileAsset.id,
+        FileShare.target_type == "campaign",
+        effective_file_share_clause(),
+    )
+    campaign_use_exists = exists().where(
+        CampaignAttachmentUse.tenant_id == tenant_id,
+        CampaignAttachmentUse.file_asset_id == FileAsset.id,
+    )
+    if campaign_usage == "linked":
+        query = query.filter(or_(campaign_share_exists, campaign_use_exists))
+    elif campaign_usage == "unlinked":
+        query = query.filter(~or_(campaign_share_exists, campaign_use_exists))
+
+    if audit_relevant is not None:
+        sent_use_exists = exists().where(
+            CampaignAttachmentUse.tenant_id == tenant_id,
+            CampaignAttachmentUse.file_asset_id == FileAsset.id,
+            CampaignAttachmentUse.use_stage == "sent",
+        )
+        query = query.filter(sent_use_exists if audit_relevant else ~sent_use_exists)
     return query
 
 
+def count_assets_for_user(
+    session: Session,
+    *,
+    tenant_id: str,
+    user_id: str,
+    owner_type: str | None = None,
+    owner_id: str | None = None,
+    campaign_id: str | None = None,
+    path_prefix: str | None = None,
+    campaign_usage: str | None = None,
+    audit_relevant: bool | None = None,
+    include_deleted: bool = False,
+    is_admin: bool = False,
+) -> int:
+    query = _asset_visibility_query_for_user(
+        session,
+        tenant_id=tenant_id,
+        user_id=user_id,
+        owner_type=owner_type,
+        owner_id=owner_id,
+        campaign_id=campaign_id,
+        path_prefix=path_prefix,
+        campaign_usage=campaign_usage,
+        audit_relevant=audit_relevant,
+        include_deleted=include_deleted,
+        is_admin=is_admin,
+    )
+    return int(
+        query.order_by(None)
+        .with_entities(func.count(func.distinct(FileAsset.id)))
+        .scalar()
+        or 0
+    )
+
+
 def current_version_and_blob(session: Session, asset: FileAsset) -> tuple[FileVersion, FileBlob]:
     if not asset.current_version_id:
         raise FileStorageError("File has no current version")
@@ -495,10 +764,22 @@ def current_versions_and_blobs(session: Session, assets: Iterable[FileAsset]) ->
 def read_asset_bytes(session: Session, asset: FileAsset) -> tuple[bytes, FileVersion, FileBlob]:
     version, blob = current_version_and_blob(session, asset)
     backend = get_storage_backend()
-    try:
+    return read_verified_blob_bytes(blob, backend=backend), version, blob
-        return backend.get_bytes(blob.storage_key), version, blob
+
-    except StorageBackendError as exc:
+
-        raise FileStorageError(str(exc)) from exc
+def read_asset_version_bytes(
+    session: Session,
+    asset: FileAsset,
+    version_id: str,
+) -> tuple[bytes, FileVersion, FileBlob]:
+    version = session.get(FileVersion, version_id)
+    if version is None or version.file_asset_id != asset.id:
+        raise FileStorageError("File version not found")
+    blob = session.get(FileBlob, version.blob_id)
+    if blob is None or blob.tenant_id != asset.tenant_id:
+        raise FileStorageError("File blob not found")
+    backend = get_storage_backend()
+    return read_verified_blob_bytes(blob, backend=backend), version, blob
 
 
 def share_file(
@@ -510,6 +791,7 @@ def share_file(
     target_id: str,
     permission: str,
     user_id: str,
+    expires_at: datetime | None = None,
 ) -> FileShare:
     target_type = target_type.lower().strip()
     permission = permission.lower().strip()
@@ -519,6 +801,7 @@ def share_file(
         raise FileStorageError("Unsupported share target")
     if permission not in {"read", "write", "manage"}:
         raise FileStorageError("Unsupported file permission")
+    _validate_share_expiry(expires_at)
     if target_type in {"user", "group", "tenant"}:
         ensure_share_target_exists(tenant_id=tenant_id, target_type=target_type, target_id=target_id)
     if target_type == "campaign":
@@ -536,6 +819,7 @@ def share_file(
     )
     if existing:
         existing.permission = permission
+        existing.expires_at = expires_at
         session.add(existing)
         return existing
     share = FileShare(
@@ -545,6 +829,7 @@ def share_file(
         target_id=target_id,
         permission=permission,
         created_by_user_id=user_id,
+        expires_at=expires_at,
     )
     session.add(share)
     return share
@@ -559,6 +844,7 @@ def share_files(
     target_id: str,
     permission: str,
     user_id: str,
+    expires_at: datetime | None = None,
 ) -> list[FileShare]:
     target_type = target_type.lower().strip()
     permission = permission.lower().strip()
@@ -566,6 +852,7 @@ def share_files(
         raise FileStorageError("Unsupported share target")
     if permission not in {"read", "write", "manage"}:
         raise FileStorageError("Unsupported file permission")
+    _validate_share_expiry(expires_at)
     if target_type in {"user", "group", "tenant"}:
         ensure_share_target_exists(tenant_id=tenant_id, target_type=target_type, target_id=target_id)
     if target_type == "campaign":
@@ -593,6 +880,7 @@ def share_files(
         existing = existing_by_asset.get(asset.id)
         if existing:
             existing.permission = permission
+            existing.expires_at = expires_at
             session.add(existing)
             shares.append(existing)
             continue
@@ -603,12 +891,88 @@ def share_files(
             target_id=target_id,
             permission=permission,
             created_by_user_id=user_id,
+            expires_at=expires_at,
         )
         session.add(share)
         shares.append(share)
     return shares
 
 
+def list_file_shares(
+    session: Session,
+    *,
+    tenant_id: str,
+    asset_id: str,
+    include_inactive: bool = False,
+) -> list[FileShare]:
+    query = session.query(FileShare).filter(
+        FileShare.tenant_id == tenant_id,
+        FileShare.file_asset_id == asset_id,
+    )
+    if not include_inactive:
+        query = query.filter(effective_file_share_clause())
+    return query.order_by(FileShare.created_at.desc(), FileShare.id.desc()).all()
+
+
+def revoke_file_share(
+    session: Session,
+    *,
+    tenant_id: str,
+    asset_id: str,
+    share_id: str,
+    user_id: str,
+) -> tuple[FileShare, bool]:
+    share = (
+        session.query(FileShare)
+        .filter(
+            FileShare.id == share_id,
+            FileShare.tenant_id == tenant_id,
+            FileShare.file_asset_id == asset_id,
+        )
+        .one_or_none()
+    )
+    if share is None:
+        raise FileStorageError("File share not found")
+    if share.revoked_at is not None:
+        return share, False
+    share.revoked_at = utcnow()
+    share.revoked_by_user_id = user_id
+    session.add(share)
+    return share, True
+
+
+def current_file_share_for_target(
+    session: Session,
+    *,
+    tenant_id: str,
+    asset_id: str,
+    target_type: str,
+    target_id: str,
+) -> FileShare | None:
+    return (
+        session.query(FileShare)
+        .filter(
+            FileShare.tenant_id == tenant_id,
+            FileShare.file_asset_id == asset_id,
+            FileShare.target_type == target_type.lower().strip(),
+            FileShare.target_id == target_id,
+            FileShare.revoked_at.is_(None),
+        )
+        .order_by(FileShare.created_at.desc(), FileShare.id.desc())
+        .first()
+    )
+
+
+def _validate_share_expiry(expires_at: datetime | None) -> None:
+    if expires_at is None:
+        return
+    from govoplan_files.backend.storage.share_state import file_share_is_active
+
+    candidate = FileShare(expires_at=expires_at)
+    if not file_share_is_active(candidate):
+        raise FileStorageError("File share expiry must be in the future")
+
+
 
 def soft_delete_assets(session: Session, assets: Iterable[FileAsset]) -> int:
     count = 0
@@ -744,7 +1108,7 @@ def _resolution_by_path(conflict_resolutions: Iterable[FileConflictResolution] |
 
 def _normalize_conflict_strategy(strategy: str | None) -> str:
     normalized = (strategy or "reject").lower().strip()
-    if normalized not in {"reject", "overwrite", "rename"}:
+    if normalized not in {"reject", "overwrite", "rename", "skip"}:
         raise FileStorageError("Unsupported conflict strategy")
     return normalized
 

src/govoplan_files/backend/storage/http_client.py

@@ -185,15 +185,23 @@ class _OutboundPolicyHTTPTransport(httpx.BaseTransport):
         self._connection_pool.close()
 
 
+_CONNECTOR_HTTP_CLIENT = httpx.Client(
+    transport=_OutboundPolicyHTTPTransport(),
+    follow_redirects=False,
+    timeout=15.0,
+)
+
+
 @contextmanager
 def _stream_connector_request(method: str, url: str, **kwargs: Any) -> Iterator[httpx.Response]:
-    with httpx.Client(
+    timeout = kwargs.pop("timeout", 15.0)
-        transport=_OutboundPolicyHTTPTransport(),
+    with _CONNECTOR_HTTP_CLIENT.stream(
-        follow_redirects=False,
+        method,
-        timeout=kwargs.pop("timeout", 15.0),
+        url,
-    ) as client:
+        timeout=timeout,
-        with client.stream(method, url, **kwargs) as response:
+        **kwargs,
-            yield response
+    ) as response:
+        yield response
 
 
 def request_connector_bytes(

src/govoplan_files/backend/storage/integrity.py

@@ -0,0 +1,550 @@
+from __future__ import annotations
+
+import hashlib
+from dataclasses import dataclass
+
+from sqlalchemy.orm import Session, object_session
+
+from govoplan_files.backend.db.models import (
+    FileBlob,
+    FileIntegrityFinding,
+    FileIntegrityScan,
+)
+from govoplan_files.backend.storage.backends import (
+    StorageBackend,
+    StorageBackendError,
+    StorageObjectMissing,
+    get_storage_backend,
+)
+from govoplan_files.backend.storage.recovery import (
+    begin_orphan_cleanup_recovery,
+)
+from govoplan_files.backend.storage.common import FileStorageError, utcnow
+
+
+TERMINAL_SCAN_STATUSES = {"completed", "cancelled"}
+QUARANTINED_BLOB_STATUSES = {
+    "missing",
+    "size_mismatch",
+    "checksum_mismatch",
+}
+
+
+@dataclass(frozen=True, slots=True)
+class BlobInspection:
+    valid: bool
+    kind: str
+    observed_size_bytes: int | None = None
+    observed_checksum_sha256: str | None = None
+
+
+@dataclass(frozen=True, slots=True)
+class IntegrityActionResult:
+    action: str
+    changed: bool
+    dry_run: bool
+    finding: FileIntegrityFinding
+    inspection: BlobInspection | None = None
+
+
+def storage_prefix_for_tenant(tenant_id: str) -> str:
+    return f"tenants/{tenant_id}/files/"
+
+
+def create_integrity_scan(
+    session: Session,
+    *,
+    tenant_id: str,
+    user_id: str,
+    verify_checksums: bool = True,
+    batch_size: int = 100,
+    backend: StorageBackend | None = None,
+) -> FileIntegrityScan:
+    active_backend = backend or get_storage_backend()
+    scan = FileIntegrityScan(
+        tenant_id=tenant_id,
+        storage_backend=active_backend.name,
+        storage_prefix=storage_prefix_for_tenant(tenant_id),
+        verify_checksums=verify_checksums,
+        batch_size=max(1, min(int(batch_size), 1000)),
+        created_by_user_id=user_id,
+    )
+    session.add(scan)
+    session.flush()
+    return scan
+
+
+def run_integrity_scan_batch(
+    session: Session,
+    scan: FileIntegrityScan,
+    *,
+    backend: StorageBackend | None = None,
+) -> FileIntegrityScan:
+    if scan.status in TERMINAL_SCAN_STATUSES:
+        return scan
+    active_backend = backend or get_storage_backend()
+    if active_backend.name != scan.storage_backend:
+        raise FileStorageError(
+            "The configured storage backend changed after this integrity scan started"
+        )
+    if scan.started_at is None:
+        scan.started_at = utcnow()
+    scan.status = "running"
+    scan.last_error = None
+    if scan.phase == "blobs":
+        _scan_blob_batch(session, scan, active_backend)
+    elif scan.phase == "objects":
+        _scan_object_batch(session, scan, active_backend)
+    else:
+        scan.phase = "completed"
+        scan.status = "completed"
+        scan.completed_at = utcnow()
+    scan.revision += 1
+    session.add(scan)
+    return scan
+
+
+def mark_integrity_scan_failed(
+    scan: FileIntegrityScan,
+    *,
+    error: Exception,
+) -> None:
+    scan.status = "failed"
+    scan.last_error = type(error).__name__[:255]
+    scan.revision += 1
+
+
+def inspect_blob(
+    blob: FileBlob,
+    *,
+    backend: StorageBackend,
+    verify_checksum: bool = True,
+) -> BlobInspection:
+    expected_size = blob.storage_size_bytes if blob.storage_size_bytes is not None else blob.size_bytes
+    expected_checksum = blob.storage_checksum_sha256 if blob.storage_checksum_sha256 is not None else blob.checksum_sha256
+    try:
+        info = backend.stat(blob.storage_key)
+    except StorageObjectMissing:
+        return BlobInspection(valid=False, kind="missing")
+    if info.size_bytes != expected_size:
+        return BlobInspection(
+            valid=False,
+            kind="size_mismatch",
+            observed_size_bytes=info.size_bytes,
+        )
+    if not verify_checksum:
+        return BlobInspection(
+            valid=True,
+            kind="verified",
+            observed_size_bytes=info.size_bytes,
+        )
+    digest = hashlib.sha256()
+    observed_size = 0
+    for chunk in backend.iter_bytes(blob.storage_key):
+        observed_size += len(chunk)
+        digest.update(chunk)
+    observed_checksum = digest.hexdigest()
+    if observed_size != expected_size:
+        return BlobInspection(
+            valid=False,
+            kind="size_mismatch",
+            observed_size_bytes=observed_size,
+            observed_checksum_sha256=observed_checksum,
+        )
+    if observed_checksum != expected_checksum:
+        return BlobInspection(
+            valid=False,
+            kind="checksum_mismatch",
+            observed_size_bytes=observed_size,
+            observed_checksum_sha256=observed_checksum,
+        )
+    return BlobInspection(
+        valid=True,
+        kind="verified",
+        observed_size_bytes=observed_size,
+        observed_checksum_sha256=observed_checksum,
+    )
+
+
+def apply_blob_inspection(
+    session: Session,
+    blob: FileBlob,
+    inspection: BlobInspection,
+    *,
+    checked_at=None,
+) -> None:
+    timestamp = checked_at or utcnow()
+    blob.integrity_checked_at = timestamp
+    if inspection.valid:
+        blob.integrity_status = "verified"
+        blob.integrity_failure = None
+        blob.quarantined_at = None
+    else:
+        blob.integrity_status = inspection.kind
+        blob.integrity_failure = inspection.kind
+        blob.quarantined_at = blob.quarantined_at or timestamp
+    session.add(blob)
+
+
+def ensure_blob_is_readable(blob: FileBlob) -> None:
+    if blob.integrity_status in QUARANTINED_BLOB_STATUSES or blob.quarantined_at:
+        raise FileStorageError(
+            "Managed file content is quarantined because integrity verification failed"
+        )
+
+
+def read_verified_blob_bytes(
+    blob: FileBlob,
+    *,
+    backend: StorageBackend,
+) -> bytes:
+    ensure_blob_is_readable(blob)
+    try:
+        data = backend.get_bytes(blob.storage_key)
+    except StorageBackendError as exc:
+        raise FileStorageError("Managed file content is not available") from exc
+    expected_storage_size = blob.storage_size_bytes if blob.storage_size_bytes is not None else blob.size_bytes
+    expected_storage_checksum = blob.storage_checksum_sha256 if blob.storage_checksum_sha256 is not None else blob.checksum_sha256
+    if len(data) != expected_storage_size:
+        raise FileStorageError(
+            "Managed file content failed its recorded size verification"
+        )
+    if hashlib.sha256(data).hexdigest() != expected_storage_checksum:
+        raise FileStorageError(
+            "Managed file content failed its recorded checksum verification"
+        )
+    if blob.encryption_envelope_id:
+        session = object_session(blob)
+        if session is None:
+            raise FileStorageError("Encrypted managed content requires an attached database session")
+        from govoplan_files.backend.storage.content_protection import unprotect_blob_content
+
+        data = unprotect_blob_content(
+            session,
+            tenant_id=blob.tenant_id,
+            blob_id=blob.id,
+            envelope_id=blob.encryption_envelope_id,
+            ciphertext=data,
+        )
+        if len(data) != blob.size_bytes:
+            raise FileStorageError("Decrypted managed file content failed its recorded size verification")
+        if hashlib.sha256(data).hexdigest() != blob.checksum_sha256:
+            raise FileStorageError("Decrypted managed file content failed its recorded checksum verification")
+    return data
+
+
+def recheck_integrity_finding(
+    session: Session,
+    finding: FileIntegrityFinding,
+    *,
+    user_id: str,
+    dry_run: bool,
+    backend: StorageBackend | None = None,
+) -> IntegrityActionResult:
+    if finding.kind == "orphan_object" or not finding.blob_id:
+        raise FileStorageError("Only blob integrity findings can be rechecked")
+    blob = session.get(FileBlob, finding.blob_id)
+    if blob is None or blob.tenant_id != finding.tenant_id:
+        raise FileStorageError("Integrity finding blob no longer exists")
+    active_backend = backend or get_storage_backend()
+    scan = session.get(FileIntegrityScan, finding.scan_id)
+    if scan is None or scan.storage_backend != active_backend.name:
+        raise FileStorageError(
+            "The configured storage backend does not match the integrity finding"
+        )
+    inspection = inspect_blob(
+        blob,
+        backend=active_backend,
+        verify_checksum=True,
+    )
+    changed = False
+    if not dry_run:
+        previous = (
+            blob.integrity_status,
+            blob.quarantined_at,
+            finding.state,
+        )
+        apply_blob_inspection(session, blob, inspection)
+        _update_finding_from_inspection(finding, inspection)
+        if inspection.valid:
+            finding.state = "resolved"
+            finding.resolved_at = utcnow()
+            finding.resolved_by_user_id = user_id
+        session.add(finding)
+        finding.revision += 1
+        changed = previous != (
+            blob.integrity_status,
+            blob.quarantined_at,
+            finding.state,
+        )
+    return IntegrityActionResult(
+        action="recheck",
+        changed=changed,
+        dry_run=dry_run,
+        finding=finding,
+        inspection=inspection,
+    )
+
+
+def cleanup_orphan_finding(
+    session: Session,
+    finding: FileIntegrityFinding,
+    *,
+    user_id: str,
+    dry_run: bool,
+    backend: StorageBackend | None = None,
+) -> IntegrityActionResult:
+    if finding.kind != "orphan_object":
+        raise FileStorageError("Only orphan-object findings can be cleaned up")
+    scan = session.get(FileIntegrityScan, finding.scan_id)
+    if scan is None or scan.tenant_id != finding.tenant_id:
+        raise FileStorageError("Integrity scan not found")
+    if not finding.storage_key.startswith(scan.storage_prefix):
+        raise FileStorageError("Orphan cleanup is outside the scan storage scope")
+    referenced = (
+        session.query(FileBlob.id)
+        .filter(
+            FileBlob.tenant_id == finding.tenant_id,
+            FileBlob.storage_backend == scan.storage_backend,
+            FileBlob.storage_key == finding.storage_key,
+        )
+        .first()
+    )
+    if referenced:
+        if not dry_run and finding.state != "resolved":
+            finding.state = "resolved"
+            finding.resolved_at = utcnow()
+            finding.resolved_by_user_id = user_id
+            finding.revision += 1
+            session.add(finding)
+            return IntegrityActionResult(
+                action="retained_referenced",
+                changed=True,
+                dry_run=False,
+                finding=finding,
+            )
+        return IntegrityActionResult(
+            action="retained_referenced",
+            changed=False,
+            dry_run=dry_run,
+            finding=finding,
+        )
+    if finding.state == "deleted":
+        return IntegrityActionResult(
+            action="already_deleted",
+            changed=False,
+            dry_run=dry_run,
+            finding=finding,
+        )
+    if dry_run:
+        return IntegrityActionResult(
+            action="would_delete",
+            changed=False,
+            dry_run=True,
+            finding=finding,
+        )
+    active_backend = backend or get_storage_backend()
+    if active_backend.name != scan.storage_backend:
+        raise FileStorageError(
+            "The configured storage backend does not match the integrity finding"
+        )
+    begin_orphan_cleanup_recovery(
+        session,
+        finding,
+        backend=active_backend,
+        user_id=user_id,
+    )
+    try:
+        active_backend.stat(finding.storage_key)
+    except StorageObjectMissing:
+        action = "already_absent"
+    else:
+        active_backend.delete(finding.storage_key)
+        action = "deleted"
+    finding.state = "deleted"
+    finding.resolved_at = utcnow()
+    finding.resolved_by_user_id = user_id
+    finding.revision += 1
+    session.add(finding)
+    return IntegrityActionResult(
+        action=action,
+        changed=True,
+        dry_run=False,
+        finding=finding,
+    )
+
+
+def _scan_blob_batch(
+    session: Session,
+    scan: FileIntegrityScan,
+    backend: StorageBackend,
+) -> None:
+    query = session.query(FileBlob).filter(
+        FileBlob.tenant_id == scan.tenant_id,
+        FileBlob.storage_backend == scan.storage_backend,
+    )
+    if scan.blob_cursor:
+        query = query.filter(FileBlob.id > scan.blob_cursor)
+    blobs = query.order_by(FileBlob.id.asc()).limit(scan.batch_size).all()
+    for blob in blobs:
+        inspection = inspect_blob(
+            blob,
+            backend=backend,
+            verify_checksum=scan.verify_checksums,
+        )
+        apply_blob_inspection(session, blob, inspection)
+        scan.scanned_blob_count += 1
+        if inspection.valid:
+            scan.verified_blob_count += 1
+            _resolve_scan_blob_findings(session, scan, blob)
+        else:
+            scan.quarantined_blob_count += 1
+            _record_blob_finding(session, scan, blob, inspection)
+        scan.blob_cursor = blob.id
+    if len(blobs) < scan.batch_size:
+        scan.phase = "objects"
+        scan.object_cursor = None
+
+
+def _scan_object_batch(
+    session: Session,
+    scan: FileIntegrityScan,
+    backend: StorageBackend,
+) -> None:
+    page = backend.list_objects(
+        prefix=scan.storage_prefix,
+        after=scan.object_cursor,
+        limit=scan.batch_size,
+    )
+    keys = [item.key for item in page.objects]
+    referenced_keys = {
+        row[0]
+        for row in session.query(FileBlob.storage_key)
+        .filter(
+            FileBlob.tenant_id == scan.tenant_id,
+            FileBlob.storage_backend == scan.storage_backend,
+            FileBlob.storage_key.in_(keys),
+        )
+        .all()
+    } if keys else set()
+    for item in page.objects:
+        scan.scanned_object_count += 1
+        if item.key in referenced_keys:
+            continue
+        scan.orphan_object_count += 1
+        _record_orphan_finding(
+            session,
+            scan,
+            storage_key=item.key,
+            size_bytes=item.size_bytes,
+        )
+    scan.object_cursor = page.next_cursor
+    if page.next_cursor is None:
+        scan.phase = "completed"
+        scan.status = "completed"
+        scan.completed_at = utcnow()
+
+
+def _record_blob_finding(
+    session: Session,
+    scan: FileIntegrityScan,
+    blob: FileBlob,
+    inspection: BlobInspection,
+) -> FileIntegrityFinding:
+    finding = (
+        session.query(FileIntegrityFinding)
+        .filter(
+            FileIntegrityFinding.scan_id == scan.id,
+            FileIntegrityFinding.blob_id == blob.id,
+            FileIntegrityFinding.kind == inspection.kind,
+        )
+        .one_or_none()
+    )
+    if finding is None:
+        finding = FileIntegrityFinding(
+            scan_id=scan.id,
+            tenant_id=scan.tenant_id,
+            kind=inspection.kind,
+            blob_id=blob.id,
+            storage_key=blob.storage_key,
+            expected_size_bytes=blob.storage_size_bytes if blob.storage_size_bytes is not None else blob.size_bytes,
+            expected_checksum_sha256=blob.storage_checksum_sha256 if blob.storage_checksum_sha256 is not None else blob.checksum_sha256,
+        )
+    else:
+        finding.revision += 1
+    _update_finding_from_inspection(finding, inspection)
+    session.add(finding)
+    return finding
+
+
+def _record_orphan_finding(
+    session: Session,
+    scan: FileIntegrityScan,
+    *,
+    storage_key: str,
+    size_bytes: int,
+) -> FileIntegrityFinding:
+    finding = (
+        session.query(FileIntegrityFinding)
+        .filter(
+            FileIntegrityFinding.scan_id == scan.id,
+            FileIntegrityFinding.kind == "orphan_object",
+            FileIntegrityFinding.storage_key == storage_key,
+        )
+        .one_or_none()
+    )
+    if finding is None:
+        finding = FileIntegrityFinding(
+            scan_id=scan.id,
+            tenant_id=scan.tenant_id,
+            kind="orphan_object",
+            storage_key=storage_key,
+            observed_size_bytes=size_bytes,
+        )
+        session.add(finding)
+    return finding
+
+
+def _resolve_scan_blob_findings(
+    session: Session,
+    scan: FileIntegrityScan,
+    blob: FileBlob,
+) -> None:
+    for finding in (
+        session.query(FileIntegrityFinding)
+        .filter(
+            FileIntegrityFinding.scan_id == scan.id,
+            FileIntegrityFinding.blob_id == blob.id,
+            FileIntegrityFinding.state == "open",
+        )
+        .all()
+    ):
+        finding.state = "resolved"
+        finding.resolved_at = utcnow()
+        finding.revision += 1
+        session.add(finding)
+
+
+def _update_finding_from_inspection(
+    finding: FileIntegrityFinding,
+    inspection: BlobInspection,
+) -> None:
+    finding.observed_size_bytes = inspection.observed_size_bytes
+    finding.observed_checksum_sha256 = inspection.observed_checksum_sha256
+
+
+__all__ = [
+    "BlobInspection",
+    "IntegrityActionResult",
+    "QUARANTINED_BLOB_STATUSES",
+    "apply_blob_inspection",
+    "cleanup_orphan_finding",
+    "create_integrity_scan",
+    "ensure_blob_is_readable",
+    "inspect_blob",
+    "mark_integrity_scan_failed",
+    "read_verified_blob_bytes",
+    "recheck_integrity_finding",
+    "run_integrity_scan_batch",
+    "storage_prefix_for_tenant",
+]

src/govoplan_files/backend/storage/lifecycle.py

@@ -0,0 +1,796 @@
+from __future__ import annotations
+
+from dataclasses import dataclass
+from datetime import UTC, datetime
+import hashlib
+import json
+from typing import Callable, Iterable
+
+from sqlalchemy import func, inspect, or_
+from sqlalchemy.orm import Session
+
+from govoplan_core.core.recovery import (
+    RecoveryGuaranteeError,
+    RecoveryMode,
+    RecoveryPlan,
+    RecoveryStatus,
+)
+from govoplan_core.core.recovery_runtime import (
+    RecoveryOperationBusy,
+    RecoveryOperationStateConflict,
+    begin_durable_recovery_operation,
+)
+from govoplan_core.core.runtime_coordination import process_runtime_identity
+from govoplan_core.db.session import get_database
+from govoplan_files.backend.db.models import (
+    CampaignAttachmentUse,
+    FileAsset,
+    FileBlob,
+    FileConnectorSpace,
+    FileFolder,
+    FileFormEvidenceGrant,
+    FileShare,
+    FileVersion,
+)
+from govoplan_files.backend.storage.access import ensure_owner_access
+from govoplan_files.backend.storage.backends import (
+    StorageBackendError,
+    get_storage_backend,
+)
+from govoplan_files.backend.storage.common import FileStorageError, utcnow
+from govoplan_files.backend.storage.connector_spaces import connector_space_owner_id
+from govoplan_files.backend.storage.paths import normalize_folder
+from govoplan_files.backend.storage.share_state import effective_file_share_clause
+
+
+@dataclass(frozen=True, slots=True)
+class PurgePreviewItem:
+    file_id: str
+    filename: str
+    lifecycle_revision: int
+    deleted_at: datetime | None
+    retained_until: datetime | None
+    legal_hold: bool
+    blockers: tuple[str, ...]
+    blob_ids: tuple[str, ...]
+
+    def digest_payload(self) -> dict[str, object]:
+        return {
+            "file_id": self.file_id,
+            "filename": self.filename,
+            "lifecycle_revision": self.lifecycle_revision,
+            "deleted_at": _iso(self.deleted_at),
+            "retained_until": _iso(self.retained_until),
+            "legal_hold": self.legal_hold,
+            "blockers": list(self.blockers),
+            "blob_ids": list(self.blob_ids),
+        }
+
+
+@dataclass(frozen=True, slots=True)
+class PurgePreview:
+    preview_sha256: str
+    items: tuple[PurgePreviewItem, ...]
+
+    @property
+    def eligible(self) -> bool:
+        return all(not item.blockers for item in self.items)
+
+
+@dataclass(frozen=True, slots=True)
+class PurgeResult:
+    recovery_operation_id: str
+    status: str
+    replayed: bool
+    purged_files: int
+    released_blobs: int
+
+
+@dataclass(frozen=True, slots=True)
+class BlobGcResult:
+    inspected_blobs: int
+    deleted_blobs: int
+    unresolved_operation_ids: tuple[str, ...]
+
+
+def set_asset_lifecycle(
+    session: Session,
+    asset: FileAsset,
+    *,
+    retained_until: datetime | None,
+    legal_hold: bool,
+    reason: str,
+    expected_revision: int,
+) -> FileAsset:
+    if asset.lifecycle_revision != expected_revision:
+        raise FileStorageError("File lifecycle settings changed; reload before saving")
+    asset.retained_until = retained_until
+    asset.legal_hold = bool(legal_hold)
+    asset.lifecycle_reason = reason.strip()
+    asset.lifecycle_revision += 1
+    session.add(asset)
+    session.flush()
+    return asset
+
+
+def get_asset_for_lifecycle(
+    session: Session,
+    *,
+    tenant_id: str,
+    user_id: str,
+    asset_id: str,
+    is_admin: bool,
+) -> FileAsset:
+    asset = session.get(FileAsset, asset_id)
+    if asset is None or asset.tenant_id != tenant_id:
+        raise FileStorageError("File not found")
+    ensure_owner_access(
+        session,
+        tenant_id=tenant_id,
+        owner_type=asset.owner_type,
+        owner_id=_asset_owner_id(asset),
+        user_id=user_id,
+        is_admin=is_admin,
+    )
+    return asset
+
+
+def restore_asset(session: Session, asset: FileAsset) -> bool:
+    if asset.deleted_at is None:
+        return False
+    collision = _asset_owner_query(session, asset).filter(
+        FileAsset.id != asset.id,
+        FileAsset.display_path == asset.display_path,
+        FileAsset.deleted_at.is_(None),
+    ).first()
+    if collision is not None:
+        raise FileStorageError(
+            "The file cannot be restored because its path is already in use"
+        )
+    asset.deleted_at = None
+    asset.lifecycle_revision += 1
+    session.add(asset)
+    session.flush()
+    return True
+
+
+def restore_folder(
+    session: Session,
+    *,
+    tenant_id: str,
+    owner_type: str,
+    owner_id: str,
+    user_id: str,
+    path: str,
+    recursive: bool,
+    is_admin: bool,
+) -> tuple[int, int]:
+    ensure_owner_access(
+        session,
+        tenant_id=tenant_id,
+        owner_type=owner_type,
+        owner_id=owner_id,
+        user_id=user_id,
+        is_admin=is_admin,
+    )
+    normalized = normalize_folder(path)
+    if not normalized:
+        raise FileStorageError("Folder path is required")
+    prefix = f"{normalized}/"
+    folders = _folder_owner_query(
+        session, tenant_id=tenant_id, owner_type=owner_type, owner_id=owner_id
+    ).filter(FileFolder.deleted_at.is_not(None))
+    assets = _asset_owner_query_by_values(
+        session, tenant_id=tenant_id, owner_type=owner_type, owner_id=owner_id
+    ).filter(FileAsset.deleted_at.is_not(None))
+    if recursive:
+        folders = folders.filter(
+            (FileFolder.path == normalized) | FileFolder.path.like(f"{prefix}%")
+        )
+        assets = assets.filter(FileAsset.display_path.like(f"{prefix}%"))
+    else:
+        folders = folders.filter(FileFolder.path == normalized)
+        assets = assets.filter(False)
+    folder_rows = folders.order_by(FileFolder.path.asc()).all()
+    asset_rows = assets.order_by(FileAsset.display_path.asc()).all()
+    if not folder_rows:
+        raise FileStorageError("Deleted folder not found")
+    for folder in folder_rows:
+        collision = _folder_owner_query(
+            session,
+            tenant_id=tenant_id,
+            owner_type=owner_type,
+            owner_id=owner_id,
+        ).filter(
+            FileFolder.id != folder.id,
+            FileFolder.path == folder.path,
+            FileFolder.deleted_at.is_(None),
+        ).first()
+        if collision is not None:
+            raise FileStorageError(
+                f"The folder cannot be restored because its path is in use: {folder.path}"
+            )
+    for asset in asset_rows:
+        collision = _asset_owner_query(session, asset).filter(
+            FileAsset.id != asset.id,
+            FileAsset.display_path == asset.display_path,
+            FileAsset.deleted_at.is_(None),
+        ).first()
+        if collision is not None:
+            raise FileStorageError(
+                f"A file cannot be restored because its path is in use: {asset.display_path}"
+            )
+    for folder in folder_rows:
+        folder.deleted_at = None
+        session.add(folder)
+    for asset in asset_rows:
+        asset.deleted_at = None
+        asset.lifecycle_revision += 1
+        session.add(asset)
+    session.flush()
+    return len(folder_rows), len(asset_rows)
+
+
+def restore_connector_space(
+    session: Session,
+    *,
+    tenant_id: str,
+    user_id: str,
+    space_id: str,
+    is_admin: bool,
+) -> FileConnectorSpace:
+    space = session.get(FileConnectorSpace, space_id)
+    if space is None or space.tenant_id != tenant_id or space.deleted_at is None:
+        raise FileStorageError("Deleted connector space not found")
+    ensure_owner_access(
+        session,
+        tenant_id=tenant_id,
+        owner_type=space.owner_type,
+        owner_id=connector_space_owner_id(space),
+        user_id=user_id,
+        is_admin=is_admin,
+    )
+    collision = _connector_space_owner_query(session, space).filter(
+        FileConnectorSpace.id != space.id,
+        FileConnectorSpace.label == space.label,
+        FileConnectorSpace.deleted_at.is_(None),
+    ).first()
+    if collision is not None:
+        raise FileStorageError(
+            "The connector space cannot be restored because its label is in use"
+        )
+    space.deleted_at = None
+    space.is_active = True
+    session.add(space)
+    session.flush()
+    return space
+
+
+def preview_asset_purge(
+    session: Session, *, tenant_id: str, file_ids: Iterable[str]
+) -> PurgePreview:
+    normalized_ids = sorted(set(str(value).strip() for value in file_ids if str(value).strip()))
+    if not normalized_ids or len(normalized_ids) > 100:
+        raise FileStorageError("Purge preview must contain between 1 and 100 files")
+    assets = (
+        session.query(FileAsset)
+        .filter(FileAsset.tenant_id == tenant_id, FileAsset.id.in_(normalized_ids))
+        .order_by(FileAsset.id.asc())
+        .all()
+    )
+    by_id = {asset.id: asset for asset in assets}
+    items: list[PurgePreviewItem] = []
+    for file_id in normalized_ids:
+        asset = by_id.get(file_id)
+        if asset is None:
+            items.append(
+                PurgePreviewItem(
+                    file_id=file_id,
+                    filename="",
+                    lifecycle_revision=0,
+                    deleted_at=None,
+                    retained_until=None,
+                    legal_hold=False,
+                    blockers=("not_found",),
+                    blob_ids=(),
+                )
+            )
+            continue
+        blockers = _asset_purge_blockers(session, asset)
+        blob_ids = tuple(
+            sorted(
+                row[0]
+                for row in session.query(FileVersion.blob_id)
+                .filter(FileVersion.file_asset_id == asset.id)
+                .distinct()
+                .all()
+            )
+        )
+        items.append(
+            PurgePreviewItem(
+                file_id=asset.id,
+                filename=asset.filename,
+                lifecycle_revision=asset.lifecycle_revision,
+                deleted_at=asset.deleted_at,
+                retained_until=asset.retained_until,
+                legal_hold=asset.legal_hold,
+                blockers=tuple(blockers),
+                blob_ids=blob_ids,
+            )
+        )
+    payload = [item.digest_payload() for item in items]
+    digest = hashlib.sha256(
+        json.dumps(payload, sort_keys=True, separators=(",", ":")).encode("utf-8")
+    ).hexdigest()
+    return PurgePreview(preview_sha256=digest, items=tuple(items))
+
+
+def execute_asset_purge(
+    session: Session,
+    *,
+    tenant_id: str,
+    file_ids: Iterable[str],
+    preview_sha256: str,
+    idempotency_key: str,
+    approval_reference: str,
+    before_commit: Callable[[list[FileAsset]], None] | None = None,
+) -> PurgeResult:
+    normalized_ids = sorted(set(str(value).strip() for value in file_ids if str(value).strip()))
+    if not normalized_ids or len(normalized_ids) > 100:
+        raise FileStorageError("Purge must contain between 1 and 100 files")
+    request = {
+        "tenant_id": tenant_id,
+        "file_ids": normalized_ids,
+        "preview_sha256": preview_sha256,
+        "approval_reference": approval_reference,
+    }
+    try:
+        started = begin_durable_recovery_operation(
+            get_database().SessionLocal,
+            identity=process_runtime_identity(),
+            module_id="files",
+            operation_type="asset-hard-purge",
+            idempotency_key=f"files-asset-purge:{idempotency_key}",
+            request=request,
+            recovery_plan=RecoveryPlan(
+                mode=RecoveryMode.IRREVERSIBLE,
+                preconditions=(
+                    "the actor holds the dedicated Files purge permission",
+                    "the signed-off preview still matches current lifecycle state",
+                    "every target is soft-deleted and free of retention or legal-hold blockers",
+                ),
+                verification_steps=(
+                    "verify every target FileAsset and FileVersion row is absent",
+                    "recalculate retained FileBlob reference counts before later garbage collection",
+                ),
+                approval_reference=approval_reference,
+            ),
+            precondition_evidence={
+                "preview_sha256": preview_sha256,
+                "target_count": len(normalized_ids),
+            },
+            lease_resource_key=f"files:purge:{tenant_id}",
+            lease_ttl_seconds=15 * 60,
+            resource_type="file_asset_batch",
+            resource_id=preview_sha256,
+            metadata={"resources": ["postgresql"], "bounded_target_count": len(normalized_ids)},
+            block_unresolved_resource=True,
+        )
+    except (RecoveryOperationBusy, RecoveryOperationStateConflict) as exc:
+        raise FileStorageError(
+            "Another purge or unresolved retry owns the tenant purge fence"
+        ) from exc
+    except (RecoveryGuaranteeError, RuntimeError, ValueError) as exc:
+        raise FileStorageError("The Files recovery ledger is unavailable; nothing was purged") from exc
+    if started.replayed or started.operation is None:
+        return PurgeResult(
+            recovery_operation_id=started.operation_id,
+            status=started.status,
+            replayed=True,
+            purged_files=0,
+            released_blobs=0,
+        )
+    operation = started.operation
+    try:
+        preview = preview_asset_purge(session, tenant_id=tenant_id, file_ids=normalized_ids)
+        if preview.preview_sha256 != preview_sha256:
+            operation.reject(
+                summary="The purge preview became stale before execution",
+                evidence=_verified_evidence(
+                    {"preview_matches": False},
+                    expected_preview_sha256=preview_sha256,
+                    observed_preview_sha256=preview.preview_sha256,
+                ),
+            )
+            raise FileStorageError("The purge preview is stale; preview again")
+        if not preview.eligible:
+            operation.reject(
+                summary="The purge was blocked by current lifecycle policy",
+                evidence=_verified_evidence(
+                    {"lifecycle_policy_allows": False},
+                    blocked_files=[
+                        {"file_id": item.file_id, "blockers": list(item.blockers)}
+                        for item in preview.items
+                        if item.blockers
+                    ]
+                ),
+            )
+            raise FileStorageError("One or more files are not eligible for purge")
+        assets = (
+            session.query(FileAsset)
+            .filter(FileAsset.tenant_id == tenant_id, FileAsset.id.in_(normalized_ids))
+            .with_for_update()
+            .all()
+        )
+        blob_ids = sorted({blob_id for item in preview.items for blob_id in item.blob_ids})
+        if before_commit is not None:
+            before_commit(assets)
+        session.query(FileShare).filter(
+            FileShare.file_asset_id.in_(normalized_ids)
+        ).delete(synchronize_session=False)
+        session.query(FileVersion).filter(
+            FileVersion.file_asset_id.in_(normalized_ids)
+        ).delete(synchronize_session=False)
+        for asset in assets:
+            session.delete(asset)
+        session.flush()
+        released_blobs = 0
+        for blob_id in blob_ids:
+            references = int(
+                session.query(func.count(FileVersion.id))
+                .filter(FileVersion.blob_id == blob_id)
+                .scalar()
+                or 0
+            )
+            blob = session.get(FileBlob, blob_id)
+            if blob is not None:
+                blob.ref_count = references
+                session.add(blob)
+                if references == 0:
+                    released_blobs += 1
+        operation.commit_verified_success(
+            session,
+            evidence=_verified_evidence(
+                {
+                    "assets_absent": True,
+                    "versions_absent": True,
+                    "blob_reference_counts_recalculated": True,
+                },
+                purged_file_ids=normalized_ids,
+                purged_files=len(assets),
+                released_blobs=released_blobs,
+                preview_sha256=preview_sha256,
+            ),
+        )
+        return PurgeResult(
+            recovery_operation_id=started.operation_id,
+            status=RecoveryStatus.SUCCEEDED.value,
+            replayed=False,
+            purged_files=len(assets),
+            released_blobs=released_blobs,
+        )
+    except FileStorageError:
+        session.rollback()
+        if not operation.closed:
+            operation.release_unresolved()
+        raise
+    except Exception as exc:
+        session.rollback()
+        if not operation.closed:
+            operation.reject(
+                summary="The purge failed before its database transaction committed",
+                evidence=_verified_evidence(
+                    {"database_transaction_committed": False},
+                    exception_type=type(exc).__name__,
+                ),
+            )
+        raise
+
+
+def garbage_collect_unreferenced_blobs(
+    session: Session,
+    *,
+    tenant_id: str,
+    limit: int,
+    approval_reference: str,
+) -> BlobGcResult:
+    candidate_rows = (
+        session.query(FileBlob)
+        .filter(FileBlob.tenant_id == tenant_id)
+        .filter(
+            or_(FileBlob.retained_until.is_(None), FileBlob.retained_until <= utcnow())
+        )
+        .filter(
+            ~session.query(FileVersion.id)
+            .filter(FileVersion.blob_id == FileBlob.id)
+            .exists()
+        )
+        .order_by(FileBlob.created_at.asc(), FileBlob.id.asc())
+        .limit(limit)
+        .all()
+    )
+    candidates = tuple(
+        (row.id, row.storage_key) for row in candidate_rows
+    )
+    # Close the read snapshot before the independent recovery transaction takes
+    # its durable SQLite/PostgreSQL write lock.
+    session.commit()
+    deleted = 0
+    unresolved: list[str] = []
+    backend = get_storage_backend()
+    for candidate_id, storage_key in candidates:
+        key_digest = hashlib.sha256(storage_key.encode("utf-8")).hexdigest()
+        request = {
+            "tenant_id": tenant_id,
+            "blob_id": candidate_id,
+            "storage_key_sha256": key_digest,
+            "approval_reference": approval_reference,
+        }
+        try:
+            started = begin_durable_recovery_operation(
+                get_database().SessionLocal,
+                identity=process_runtime_identity(),
+                module_id="files",
+                operation_type="blob-garbage-collection",
+                idempotency_key=(
+                    f"files-blob-gc:{candidate_id}:{key_digest[:12]}:"
+                    f"{hashlib.sha256(approval_reference.encode('utf-8')).hexdigest()[:12]}"
+                ),
+                request=request,
+                recovery_plan=RecoveryPlan(
+                    mode=RecoveryMode.FORWARD_RECOVERY,
+                    preconditions=(
+                        "the caller holds dedicated Files purge authority",
+                        "a fresh database check shows no FileVersion reference",
+                    ),
+                    forward_recovery_steps=(
+                        "verify the object is absent",
+                        "delete the unreferenced FileBlob row after absence is proven",
+                    ),
+                    verification_steps=(
+                        "recheck FileVersion references while holding the blob fence",
+                        "probe the exact managed storage key after deletion",
+                    ),
+                    approval_reference=approval_reference,
+                ),
+                precondition_evidence={
+                    "blob_id": candidate_id,
+                    "storage_key_sha256": key_digest,
+                    "observed_reference_count": 0,
+                },
+                lease_resource_key=f"files:blob:{tenant_id}:{candidate_id}",
+                lease_ttl_seconds=15 * 60,
+                resource_type="file_blob",
+                resource_id=candidate_id,
+                metadata={"resources": ["postgresql", "object-storage"], "storage_backend": backend.name},
+                block_unresolved_resource=True,
+            )
+        except (RecoveryOperationBusy, RecoveryOperationStateConflict):
+            continue
+        except (RecoveryGuaranteeError, RuntimeError, ValueError) as exc:
+            raise FileStorageError(
+                "The Files recovery ledger is unavailable; no blob bytes were deleted"
+            ) from exc
+        if started.replayed or started.operation is None:
+            continue
+        operation = started.operation
+        try:
+            blob = (
+                session.query(FileBlob)
+                .filter(FileBlob.id == candidate_id, FileBlob.tenant_id == tenant_id)
+                .with_for_update()
+                .one_or_none()
+            )
+            references = int(
+                session.query(func.count(FileVersion.id))
+                .filter(FileVersion.blob_id == candidate_id)
+                .scalar()
+                or 0
+            )
+            blob_retained = bool(
+                blob is not None
+                and blob.retained_until is not None
+                and _as_utc(blob.retained_until) > utcnow()
+            )
+        except Exception as exc:
+            session.rollback()
+            if not operation.closed:
+                operation.reject(
+                    summary="Blob eligibility could not be rechecked before deletion",
+                    evidence=_verified_evidence(
+                        {"external_effect_started": False},
+                        exception_type=type(exc).__name__,
+                    ),
+                )
+            raise FileStorageError(
+                "Blob eligibility could not be rechecked; no bytes were deleted"
+            ) from exc
+        if blob is None or references or blob_retained:
+            operation.reject(
+                summary="Blob garbage collection was no longer eligible",
+                evidence=_verified_evidence(
+                    {
+                        "blob_present": blob is not None,
+                        "reference_count_zero": references == 0,
+                        "blob_retention_expired": not blob_retained,
+                    },
+                    reference_count=references,
+                ),
+            )
+            session.rollback()
+            continue
+        try:
+            backend.delete(blob.storage_key)
+            object_present = backend.exists(blob.storage_key)
+        except (StorageBackendError, OSError) as exc:
+            try:
+                object_present = backend.exists(blob.storage_key)
+            except (StorageBackendError, OSError):
+                object_present = None
+            session.rollback()
+            if object_present is True:
+                operation.fail(
+                    summary="The unreferenced blob object was retained",
+                    evidence={"object_present": True, "exception_type": type(exc).__name__},
+                )
+            else:
+                operation.unresolved(
+                    status=RecoveryStatus.OUTCOME_UNKNOWN,
+                    summary="The blob deletion outcome could not be verified",
+                    evidence={"object_present": object_present, "exception_type": type(exc).__name__},
+                    failure_summary="Reconcile the exact blob key before retrying garbage collection",
+                )
+                unresolved.append(started.operation_id)
+            continue
+        if object_present:
+            session.rollback()
+            operation.fail(
+                summary="The object store did not delete the unreferenced blob",
+                evidence={"object_present": True},
+            )
+            continue
+        try:
+            session.delete(blob)
+            session.flush()
+            operation.commit_verified_success(
+                session,
+                evidence=_verified_evidence(
+                    {
+                        "object_absent": True,
+                        "database_blob_absent": True,
+                        "reference_count_zero": True,
+                    },
+                    blob_id=candidate_id,
+                    storage_key_sha256=key_digest,
+                    object_present=False,
+                    database_blob_present=False,
+                    reference_count=0,
+                ),
+            )
+            deleted += 1
+        except Exception as exc:
+            session.rollback()
+            if not operation.closed:
+                operation.unresolved(
+                    status=RecoveryStatus.RECOVERY_REQUIRED,
+                    summary="The object is absent but blob metadata still requires forward recovery",
+                    evidence={"object_present": False, "exception_type": type(exc).__name__},
+                    failure_summary="Delete the FileBlob row only after rechecking all FileVersion references",
+                )
+            unresolved.append(started.operation_id)
+    return BlobGcResult(
+        inspected_blobs=len(candidates),
+        deleted_blobs=deleted,
+        unresolved_operation_ids=tuple(unresolved),
+    )
+
+
+def _asset_purge_blockers(session: Session, asset: FileAsset) -> list[str]:
+    blockers: list[str] = []
+    if asset.deleted_at is None:
+        blockers.append("not_soft_deleted")
+    if asset.legal_hold:
+        blockers.append("legal_hold")
+    if asset.retained_until is not None and _as_utc(asset.retained_until) > utcnow():
+        blockers.append("retention_active")
+    if _table_exists(session, CampaignAttachmentUse.__tablename__) and session.query(
+        CampaignAttachmentUse.id
+    ).filter(CampaignAttachmentUse.file_asset_id == asset.id).first() is not None:
+        blockers.append("campaign_evidence")
+    if _table_exists(session, FileFormEvidenceGrant.__tablename__) and session.query(
+        FileFormEvidenceGrant.id
+    ).filter(FileFormEvidenceGrant.file_asset_id == asset.id).first() is not None:
+        blockers.append("form_evidence")
+    if session.query(FileShare.id).filter(
+        FileShare.file_asset_id == asset.id,
+        effective_file_share_clause(),
+    ).first() is not None:
+        blockers.append("active_share")
+    return blockers
+
+
+def _asset_owner_id(asset: FileAsset) -> str:
+    owner_id = asset.owner_user_id if asset.owner_type == "user" else asset.owner_group_id
+    if not owner_id:
+        raise FileStorageError("File has no valid owner")
+    return owner_id
+
+
+def _asset_owner_query(session: Session, asset: FileAsset):
+    return _asset_owner_query_by_values(
+        session,
+        tenant_id=asset.tenant_id,
+        owner_type=asset.owner_type,
+        owner_id=_asset_owner_id(asset),
+    )
+
+
+def _asset_owner_query_by_values(
+    session: Session, *, tenant_id: str, owner_type: str, owner_id: str
+):
+    query = session.query(FileAsset).filter(
+        FileAsset.tenant_id == tenant_id, FileAsset.owner_type == owner_type
+    )
+    if owner_type == "user":
+        return query.filter(FileAsset.owner_user_id == owner_id)
+    if owner_type == "group":
+        return query.filter(FileAsset.owner_group_id == owner_id)
+    raise FileStorageError("Files must be owned by a user or group")
+
+
+def _folder_owner_query(
+    session: Session, *, tenant_id: str, owner_type: str, owner_id: str
+):
+    query = session.query(FileFolder).filter(
+        FileFolder.tenant_id == tenant_id, FileFolder.owner_type == owner_type
+    )
+    if owner_type == "user":
+        return query.filter(FileFolder.owner_user_id == owner_id)
+    if owner_type == "group":
+        return query.filter(FileFolder.owner_group_id == owner_id)
+    raise FileStorageError("Folders must be owned by a user or group")
+
+
+def _connector_space_owner_query(session: Session, space: FileConnectorSpace):
+    query = session.query(FileConnectorSpace).filter(
+        FileConnectorSpace.tenant_id == space.tenant_id,
+        FileConnectorSpace.owner_type == space.owner_type,
+    )
+    owner_id = connector_space_owner_id(space)
+    if space.owner_type == "user":
+        return query.filter(FileConnectorSpace.owner_user_id == owner_id)
+    return query.filter(FileConnectorSpace.owner_group_id == owner_id)
+
+
+def _as_utc(value: datetime) -> datetime:
+    return value.replace(tzinfo=UTC) if value.tzinfo is None else value.astimezone(UTC)
+
+
+def _iso(value: datetime | None) -> str | None:
+    return _as_utc(value).isoformat() if value is not None else None
+
+
+def _table_exists(session: Session, table_name: str) -> bool:
+    return bool(session.bind is not None and inspect(session.bind).has_table(table_name))
+
+
+def _verified_evidence(
+    checks: dict[str, object], **details: object
+) -> dict[str, object]:
+    return {"verified": True, "checks": checks, **details}
+
+
+__all__ = [
+    "BlobGcResult",
+    "PurgePreview",
+    "PurgePreviewItem",
+    "PurgeResult",
+    "execute_asset_purge",
+    "garbage_collect_unreferenced_blobs",
+    "get_asset_for_lifecycle",
+    "preview_asset_purge",
+    "restore_asset",
+    "restore_connector_space",
+    "restore_folder",
+    "set_asset_lifecycle",
+]

src/govoplan_files/backend/storage/recovery.py

@@ -0,0 +1,931 @@
+from __future__ import annotations
+
+from collections.abc import Callable
+from dataclasses import dataclass
+from datetime import UTC, datetime
+import hashlib
+from threading import Lock
+from typing import Protocol
+
+from sqlalchemy import event, select
+from sqlalchemy.orm import Session
+
+from govoplan_core.core.recovery import (
+    RecoveryGuaranteeError,
+    RecoveryMode,
+    RecoveryOperation,
+    RecoveryPlan,
+    RecoveryStatus,
+    plan_recovery_operation,
+    prepare_recovery_operation,
+    start_recovery_operation,
+    verify_recovery_evidence_chain,
+)
+from govoplan_core.core.recovery_runtime import (
+    DurableRecoveryOperation,
+    DurableRecoveryStart,
+    RecoveryOperationBusy,
+    RecoveryOperationStateConflict,
+    begin_durable_recovery_operation,
+)
+from govoplan_core.core.runtime_coordination import (
+    RuntimeIdentity,
+    acquire_lease,
+    process_runtime_identity,
+    release_lease,
+)
+from govoplan_core.db.session import get_database
+from govoplan_files.backend.db.models import (
+    FileBlob,
+    FileIntegrityFinding,
+    FileIntegrityScan,
+)
+from govoplan_files.backend.storage.backends import (
+    StorageBackend,
+    StorageBackendError,
+    StorageObjectMissing,
+)
+from govoplan_files.backend.storage.common import FileStorageError
+
+
+_PENDING_EFFECTS_KEY = "govoplan_files_pending_recovery_effects"
+_HOOKS_INSTALLED_KEY = "govoplan_files_recovery_hooks_installed"
+_ROLLBACK_ERRORS_KEY = "govoplan_files_recovery_rollback_errors"
+_SQLITE_FENCES: set[str] = set()
+_SQLITE_FENCES_LOCK = Lock()
+
+
+class _PendingEffect(Protocol):
+    def settle(self, *, committed: bool) -> None: ...
+
+
+@dataclass(slots=True)
+class PendingBlobWrite:
+    operation: DurableRecoveryOperation
+    backend: StorageBackend
+    tenant_id: str
+    blob_id: str
+    storage_key: str
+    semantic_checksum_sha256: str
+    semantic_size_bytes: int
+    protection_discriminator: str
+    created_new: bool
+    expected_storage_checksum_sha256: str | None = None
+    expected_storage_size_bytes: int | None = None
+    expected_envelope_id: str | None = None
+    sqlite_fence_key: str | None = None
+    restart_after_rollback: Callable[[], DurableRecoveryOperation] | None = None
+
+    def prepare_stored_bytes(
+        self,
+        data: bytes,
+        *,
+        envelope_id: str | None,
+    ) -> None:
+        """Retain process-local verification evidence without recording content."""
+
+        self.expected_storage_checksum_sha256 = hashlib.sha256(data).hexdigest()
+        self.expected_storage_size_bytes = len(data)
+        self.expected_envelope_id = envelope_id
+
+    def settle(self, *, committed: bool) -> None:
+        if not committed and self.restart_after_rollback is not None:
+            try:
+                self.operation = self.restart_after_rollback()
+            except Exception:
+                # SQLite cannot make the pre-effect ledger row independent of
+                # the caller transaction. If reconstructing the evidence row
+                # also fails, still avoid leaving a known new orphan behind.
+                if self.created_new:
+                    try:
+                        self.backend.delete(self.storage_key)
+                    except (StorageBackendError, OSError):
+                        pass
+                raise
+        evidence = _blob_write_evidence(self)
+        if _blob_write_complete(evidence):
+            self.operation.succeed(evidence=evidence)
+            return
+
+        if self.created_new and evidence.get("database_blob_present") is False:
+            self._settle_unreferenced_new_object(evidence)
+            return
+
+        if not self.created_new and _object_matches(evidence):
+            if _forward_complete_blob_repair(self):
+                completed = _blob_write_evidence(self)
+                if _blob_write_complete(completed):
+                    self.operation.succeed(evidence=completed)
+                    return
+
+        if evidence.get("database_blob_present") is True and not _object_matches(
+            evidence
+        ):
+            _quarantine_blob_after_failed_verification(self, evidence)
+            evidence = _blob_write_evidence(self)
+
+        status = (
+            RecoveryStatus.OUTCOME_UNKNOWN
+            if not evidence.get("verified")
+            else RecoveryStatus.RECOVERY_REQUIRED
+        )
+        transaction = "committed" if committed else "rolled back"
+        self.operation.unresolved(
+            status=status,
+            summary=f"Managed Files blob write remained unresolved after the database transaction {transaction}",
+            evidence=evidence,
+            failure_summary=(
+                "The managed object and Files blob metadata require reconciliation"
+            ),
+        )
+
+    def _settle_unreferenced_new_object(self, evidence: dict[str, object]) -> None:
+        object_present = evidence.get("object_present")
+        if object_present is False:
+            self.operation.reject(
+                summary="The Files blob write left no durable object or database row",
+                evidence=evidence,
+            )
+            return
+        if object_present is not True:
+            self.operation.unresolved(
+                status=RecoveryStatus.OUTCOME_UNKNOWN,
+                summary="The unreferenced Files object could not be probed",
+                evidence=evidence,
+                failure_summary="Object storage availability prevented upload compensation",
+            )
+            return
+        try:
+            self.backend.delete(self.storage_key)
+        except (StorageBackendError, OSError):
+            self.operation.unresolved(
+                status=RecoveryStatus.RECOVERY_REQUIRED,
+                summary="An unreferenced Files object could not be compensated",
+                evidence=evidence,
+                failure_summary="Delete the unreferenced managed object after verifying that no FileBlob references it",
+            )
+            return
+        recovered = _blob_write_evidence(self)
+        if (
+            recovered.get("verified") is True
+            and recovered.get("database_blob_present") is False
+            and recovered.get("object_present") is False
+        ):
+            self.operation.compensate(
+                failure_summary="The Files database transaction did not retain the new blob",
+                failure_evidence=evidence,
+                recovery_evidence=recovered,
+            )
+            return
+        self.operation.unresolved(
+            status=RecoveryStatus.RECOVERY_REQUIRED,
+            summary="Files upload compensation could not be verified",
+            evidence=recovered,
+            failure_summary="The unreferenced managed object requires operator reconciliation",
+        )
+
+
+@dataclass(slots=True)
+class PendingOrphanCleanup:
+    operation: DurableRecoveryOperation
+    backend: StorageBackend
+    finding_id: str
+    tenant_id: str
+    storage_key: str
+    resolved_by_user_id: str
+
+    def settle(self, *, committed: bool) -> None:
+        del committed
+        evidence = _orphan_cleanup_evidence(self)
+        if _orphan_cleanup_complete(evidence):
+            self.operation.succeed(evidence=evidence)
+            return
+        if (
+            evidence.get("verified") is True
+            and evidence.get("object_present") is True
+            and evidence.get("finding_deleted") is False
+        ):
+            self.operation.reject(
+                summary="The orphan object was retained and the cleanup finding stayed open",
+                evidence=evidence,
+            )
+            return
+        if evidence.get("object_present") is False:
+            if _forward_complete_orphan_finding(self):
+                completed = _orphan_cleanup_evidence(self)
+                if _orphan_cleanup_complete(completed):
+                    self.operation.succeed(evidence=completed)
+                    return
+        status = (
+            RecoveryStatus.OUTCOME_UNKNOWN
+            if not evidence.get("verified")
+            else RecoveryStatus.RECOVERY_REQUIRED
+        )
+        self.operation.unresolved(
+            status=status,
+            summary="Files orphan cleanup requires reconciliation",
+            evidence=evidence,
+            failure_summary="Recheck the object and integrity-finding state before another cleanup attempt",
+        )
+
+
+def begin_blob_write_recovery(
+    session: Session,
+    *,
+    backend: StorageBackend,
+    tenant_id: str,
+    blob_id: str,
+    storage_key: str,
+    semantic_checksum_sha256: str,
+    semantic_size_bytes: int,
+    protection_discriminator: str,
+    created_new: bool,
+    repair_token: str | None = None,
+) -> PendingBlobWrite:
+    disposition = "create" if created_new else "repair"
+    state_token = repair_token or blob_id
+    key_digest = hashlib.sha256(storage_key.encode("utf-8")).hexdigest()
+    idempotency_key = (
+        f"files-blob-{disposition}:{blob_id}:{state_token[:48]}"
+    )
+    try:
+        identity = process_runtime_identity()
+    except RuntimeError as exc:
+        raise FileStorageError(
+            "The Files recovery ledger is unavailable; no object was written"
+        ) from exc
+    lease_resource_key = f"files:blob:{tenant_id}:{blob_id}"
+    request = {
+        "tenant_id": tenant_id,
+        "blob_id": blob_id,
+        "storage_key": storage_key if created_new else None,
+        "storage_key_sha256": key_digest,
+        "semantic_checksum_sha256": semantic_checksum_sha256,
+        "semantic_size_bytes": semantic_size_bytes,
+        "protection_discriminator": protection_discriminator,
+        "disposition": disposition,
+    }
+    recovery_plan = RecoveryPlan(
+        mode=(
+            RecoveryMode.COMPENSATION
+            if created_new
+            else RecoveryMode.FORWARD_RECOVERY
+        ),
+        preconditions=(
+            "the caller has Files write authority for the target owner",
+            "the storage key belongs to the tenant Files namespace",
+            "the request records digests rather than file contents",
+        ),
+        compensation_steps=(
+            "verify that no FileBlob references the newly reserved key",
+            "delete only that unreferenced key and verify absence",
+        )
+        if created_new
+        else (),
+        forward_recovery_steps=(
+            "verify the expected bytes at the existing blob key",
+            "forward-complete matching integrity metadata or quarantine the blob",
+        )
+        if not created_new
+        else (),
+        verification_steps=(
+            "reload FileBlob metadata through an independent session",
+            "stream and hash the managed object independently",
+        ),
+    )
+    precondition_evidence = {
+        "blob_id": blob_id,
+        "storage_key_sha256": key_digest,
+        "semantic_checksum_sha256": semantic_checksum_sha256,
+        "semantic_size_bytes": semantic_size_bytes,
+        "created_new": created_new,
+    }
+    metadata = {
+        "resources": ["postgresql", "object-storage"],
+        "storage_backend": backend.name,
+        "durability_mode": "independent_transaction",
+    }
+    session_factory = get_database().SessionLocal
+    sqlite_mode = session.get_bind().dialect.name == "sqlite"
+    sqlite_fence_key: str | None = None
+
+    def start_independent(*, reconstructed_after_rollback: bool = False) -> DurableRecoveryStart:
+        return begin_durable_recovery_operation(
+            session_factory,
+            identity=identity,
+            module_id="files",
+            operation_type=f"blob-{disposition}",
+            idempotency_key=idempotency_key,
+            request=request,
+            recovery_plan=recovery_plan,
+            precondition_evidence={
+                **precondition_evidence,
+                **(
+                    {
+                        "sqlite_caller_transaction_rolled_back": True,
+                        "effect_may_have_preceded_durable_intent": True,
+                    }
+                    if reconstructed_after_rollback
+                    else {}
+                ),
+            },
+            lease_resource_key=lease_resource_key,
+            lease_ttl_seconds=15 * 60,
+            resource_type="file_blob",
+            resource_id=blob_id,
+            metadata={
+                **metadata,
+                "durability_mode": (
+                    "sqlite_post_rollback_reconstruction"
+                    if reconstructed_after_rollback
+                    else "independent_transaction"
+                ),
+            },
+        )
+
+    try:
+        if sqlite_mode:
+            _reserve_sqlite_fence(lease_resource_key)
+            sqlite_fence_key = lease_resource_key
+            started = _begin_caller_transaction_recovery_operation(
+                session,
+                session_factory=session_factory,
+                identity=identity,
+                module_id="files",
+                operation_type=f"blob-{disposition}",
+                idempotency_key=idempotency_key,
+                request=request,
+                recovery_plan=recovery_plan,
+                precondition_evidence={
+                    **precondition_evidence,
+                    "sqlite_caller_transaction": True,
+                    "reduced_crash_durability": True,
+                },
+                lease_resource_key=lease_resource_key,
+                lease_ttl_seconds=15 * 60,
+                resource_type="file_blob",
+                resource_id=blob_id,
+                metadata={
+                    **metadata,
+                    "resources": ["sqlite", "object-storage"],
+                    "durability_mode": "sqlite_caller_transaction",
+                },
+            )
+        else:
+            started = start_independent()
+    except (RecoveryOperationBusy, RecoveryOperationStateConflict) as exc:
+        if sqlite_fence_key is not None:
+            _release_sqlite_fence(sqlite_fence_key)
+        raise FileStorageError(
+            "This managed blob is already owned by another recovery operation"
+        ) from exc
+    except (RecoveryGuaranteeError, RuntimeError) as exc:
+        if sqlite_fence_key is not None:
+            _release_sqlite_fence(sqlite_fence_key)
+        raise FileStorageError(
+            "The Files recovery ledger is unavailable; no object was written"
+        ) from exc
+    if started.replayed or started.operation is None:
+        if sqlite_fence_key is not None:
+            _release_sqlite_fence(sqlite_fence_key)
+        raise FileStorageError(
+            "The matching Files blob operation was already completed; reload before retrying"
+        )
+    pending = PendingBlobWrite(
+        operation=started.operation,
+        backend=backend,
+        tenant_id=tenant_id,
+        blob_id=blob_id,
+        storage_key=storage_key,
+        semantic_checksum_sha256=semantic_checksum_sha256,
+        semantic_size_bytes=semantic_size_bytes,
+        protection_discriminator=protection_discriminator,
+        created_new=created_new,
+        sqlite_fence_key=sqlite_fence_key,
+        restart_after_rollback=(
+            lambda: _require_started_operation(
+                start_independent(reconstructed_after_rollback=True)
+            )
+            if sqlite_mode
+            else None
+        ),
+    )
+    _register_pending_effect(session, pending)
+    return pending
+
+
+def begin_orphan_cleanup_recovery(
+    session: Session,
+    finding: FileIntegrityFinding,
+    *,
+    backend: StorageBackend,
+    user_id: str,
+) -> PendingOrphanCleanup:
+    key_digest = hashlib.sha256(finding.storage_key.encode("utf-8")).hexdigest()
+    try:
+        started = begin_durable_recovery_operation(
+            get_database().SessionLocal,
+            identity=process_runtime_identity(),
+            module_id="files",
+            operation_type="integrity-orphan-cleanup",
+            idempotency_key=f"files-orphan-cleanup:{finding.id}",
+            request={
+                "tenant_id": finding.tenant_id,
+                "finding_id": finding.id,
+                "storage_key_sha256": key_digest,
+            },
+            recovery_plan=RecoveryPlan(
+                mode=RecoveryMode.FORWARD_RECOVERY,
+                preconditions=(
+                    "the integrity finding identifies an unreferenced object",
+                    "the object key remains inside the completed scan scope",
+                    "a fresh database reference check found no FileBlob",
+                ),
+                forward_recovery_steps=(
+                    "verify object absence independently",
+                    "mark the durable finding deleted only after absence is proven",
+                ),
+                verification_steps=(
+                    "reload the finding and its scan through an independent session",
+                    "probe the original storage key through the configured backend",
+                ),
+            ),
+            precondition_evidence={
+                "finding_id": finding.id,
+                "scan_id": finding.scan_id,
+                "storage_key_sha256": key_digest,
+                "finding_state": finding.state,
+            },
+            lease_resource_key=(
+                f"files:orphan-cleanup:{finding.tenant_id}:{key_digest[:40]}"
+            ),
+            lease_ttl_seconds=15 * 60,
+            resource_type="file_integrity_finding",
+            resource_id=finding.id,
+            metadata={
+                "resources": ["postgresql", "object-storage"],
+                "storage_backend": backend.name,
+            },
+        )
+    except (RecoveryOperationBusy, RecoveryOperationStateConflict) as exc:
+        raise FileStorageError(
+            "This orphan object is already owned by another recovery operation"
+        ) from exc
+    except (RecoveryGuaranteeError, RuntimeError) as exc:
+        raise FileStorageError(
+            "The Files recovery ledger is unavailable; no object was deleted"
+        ) from exc
+    if started.replayed or started.operation is None:
+        raise FileStorageError(
+            "This orphan cleanup was already completed; reload the finding"
+        )
+    pending = PendingOrphanCleanup(
+        operation=started.operation,
+        backend=backend,
+        finding_id=finding.id,
+        tenant_id=finding.tenant_id,
+        storage_key=finding.storage_key,
+        resolved_by_user_id=user_id,
+    )
+    _register_pending_effect(session, pending)
+    return pending
+
+
+def _register_pending_effect(session: Session, effect: _PendingEffect) -> None:
+    if not session.in_transaction():
+        session.begin()
+    pending = session.info.setdefault(_PENDING_EFFECTS_KEY, [])
+    pending.append(effect)
+    if session.info.get(_HOOKS_INSTALLED_KEY):
+        return
+    event.listen(session, "after_commit", _after_session_commit)
+    event.listen(session, "after_rollback", _after_session_rollback)
+    session.info[_HOOKS_INSTALLED_KEY] = True
+
+
+def _after_session_commit(session: Session) -> None:
+    # SQLAlchemy also emits after_commit for a released SAVEPOINT. A Files
+    # batch may open one while creating a later recovery row; settle only when
+    # the outer business transaction has actually become visible.
+    if session.in_nested_transaction():
+        return
+    _settle_pending_effects(session, committed=True)
+
+
+def _after_session_rollback(session: Session) -> None:
+    if session.in_nested_transaction():
+        return
+    try:
+        _settle_pending_effects(session, committed=False)
+    except RecoveryGuaranteeError as exc:
+        session.info.setdefault(_ROLLBACK_ERRORS_KEY, []).append(str(exc))
+
+
+def _settle_pending_effects(session: Session, *, committed: bool) -> None:
+    pending = list(session.info.pop(_PENDING_EFFECTS_KEY, []))
+    failures: list[Exception] = []
+    for effect in pending:
+        try:
+            effect.settle(committed=committed)
+        except Exception as exc:  # preserve every effect's chance to settle
+            failures.append(exc)
+            operation = getattr(effect, "operation", None)
+            if operation is not None and not operation.closed:
+                try:
+                    operation.release_unresolved()
+                except Exception:
+                    pass
+        finally:
+            sqlite_fence_key = getattr(effect, "sqlite_fence_key", None)
+            if sqlite_fence_key:
+                _release_sqlite_fence(sqlite_fence_key)
+    if failures:
+        raise RecoveryGuaranteeError(
+            f"{len(failures)} Files recovery operation(s) could not be finalized"
+        ) from failures[0]
+
+
+def _reserve_sqlite_fence(resource_key: str) -> None:
+    with _SQLITE_FENCES_LOCK:
+        if resource_key in _SQLITE_FENCES:
+            raise RecoveryOperationBusy(
+                f"Another local SQLite transaction owns {resource_key}"
+            )
+        _SQLITE_FENCES.add(resource_key)
+
+
+def _release_sqlite_fence(resource_key: str) -> None:
+    with _SQLITE_FENCES_LOCK:
+        _SQLITE_FENCES.discard(resource_key)
+
+
+def _require_started_operation(started: DurableRecoveryStart) -> DurableRecoveryOperation:
+    if started.replayed or started.operation is None:
+        raise RecoveryOperationStateConflict(started.operation_id, started.status)
+    return started.operation
+
+
+def _begin_caller_transaction_recovery_operation(
+    session: Session,
+    *,
+    session_factory: Callable[[], Session],
+    identity: RuntimeIdentity,
+    module_id: str,
+    operation_type: str,
+    idempotency_key: str,
+    request: dict[str, object],
+    recovery_plan: RecoveryPlan,
+    precondition_evidence: dict[str, object],
+    lease_resource_key: str,
+    lease_ttl_seconds: int,
+    resource_type: str,
+    resource_id: str,
+    metadata: dict[str, object],
+) -> DurableRecoveryStart:
+    """Start SQLite recovery evidence inside the caller transaction.
+
+    SQLite permits only one writer, so opening the normal independent ledger
+    transaction after an earlier member of a batch has written will deadlock
+    until the busy timeout. The caller-transaction mode preserves fencing,
+    request hashing, checkpoints, commit verification, and rollback
+    compensation, but cannot make the intent survive a hard process loss
+    before the caller commits. PostgreSQL never uses this reduced mode.
+    """
+
+    claim = acquire_lease(
+        session,
+        installation_id=identity.installation_id,
+        resource_key=lease_resource_key,
+        holder_node_id=identity.node_id,
+        holder_incarnation=identity.incarnation,
+        ttl_seconds=lease_ttl_seconds,
+        metadata={
+            "module_id": module_id,
+            "operation_type": operation_type,
+            "durability_mode": "sqlite_caller_transaction",
+        },
+    )
+    if claim is None:
+        raise RecoveryOperationBusy(
+            f"Another runtime owns the recovery fence for {lease_resource_key}"
+        )
+    existing = session.execute(
+        select(RecoveryOperation).where(
+            RecoveryOperation.installation_id == identity.installation_id,
+            RecoveryOperation.module_id == module_id,
+            RecoveryOperation.idempotency_key == idempotency_key,
+        )
+    ).scalar_one_or_none()
+    operation = plan_recovery_operation(
+        session,
+        installation_id=identity.installation_id,
+        module_id=module_id,
+        operation_type=operation_type,
+        idempotency_key=idempotency_key,
+        request=request,
+        recovery_plan=recovery_plan,
+        resource_type=resource_type,
+        resource_id=resource_id,
+        lease_claim=claim,
+        metadata=metadata,
+    )
+    if existing is not None:
+        if operation.status == RecoveryStatus.SUCCEEDED.value:
+            release_lease(session, claim)
+            return DurableRecoveryStart(
+                operation_id=operation.id,
+                status=operation.status,
+                replayed=True,
+                operation=None,
+            )
+        raise RecoveryOperationStateConflict(operation.id, operation.status)
+    prepare_recovery_operation(
+        session,
+        operation,
+        evidence=precondition_evidence,
+        lease_claim=claim,
+    )
+    start_recovery_operation(
+        session,
+        operation,
+        evidence={"lease_resource_key": lease_resource_key},
+        lease_claim=claim,
+    )
+    if not verify_recovery_evidence_chain(session, operation.id):
+        raise RecoveryGuaranteeError(
+            "Recovery checkpoint chain verification failed before side effects"
+        )
+    return DurableRecoveryStart(
+        operation_id=operation.id,
+        status=operation.status,
+        replayed=False,
+        operation=DurableRecoveryOperation(
+            session_factory=session_factory,
+            operation_id=operation.id,
+            lease_claim=claim,
+            lease_ttl_seconds=lease_ttl_seconds,
+        ),
+    )
+
+
+def _blob_write_evidence(effect: PendingBlobWrite) -> dict[str, object]:
+    database_blob_present: bool | None
+    database_matches: bool | None
+    database_integrity_verified: bool | None
+    try:
+        with effect.operation.session_factory() as session:
+            blob = session.get(FileBlob, effect.blob_id)
+            database_blob_present = blob is not None
+            database_matches = bool(
+                blob is not None
+                and blob.tenant_id == effect.tenant_id
+                and blob.storage_key == effect.storage_key
+                and blob.checksum_sha256 == effect.semantic_checksum_sha256
+                and blob.size_bytes == effect.semantic_size_bytes
+                and blob.protection_discriminator
+                == effect.protection_discriminator
+                and blob.encryption_envelope_id == effect.expected_envelope_id
+                and (
+                    effect.expected_storage_checksum_sha256
+                    == (
+                        blob.storage_checksum_sha256
+                        or blob.checksum_sha256
+                    )
+                )
+            ) if blob is not None else False
+            database_integrity_verified = bool(
+                blob is not None
+                and blob.integrity_status == "verified"
+                and blob.quarantined_at is None
+            ) if blob is not None else False
+    except Exception:
+        database_blob_present = None
+        database_matches = None
+        database_integrity_verified = None
+
+    object_present: bool | None
+    observed_size: int | None = None
+    observed_checksum: str | None = None
+    try:
+        digest = hashlib.sha256()
+        observed_size = 0
+        for chunk in effect.backend.iter_bytes(effect.storage_key):
+            observed_size += len(chunk)
+            digest.update(chunk)
+        observed_checksum = digest.hexdigest()
+        object_present = True
+    except StorageObjectMissing:
+        object_present = False
+    except (StorageBackendError, OSError):
+        object_present = None
+
+    object_size_matches = (
+        observed_size == effect.expected_storage_size_bytes
+        if object_present is True and effect.expected_storage_size_bytes is not None
+        else False if object_present is False else None
+    )
+    object_checksum_matches = (
+        observed_checksum == effect.expected_storage_checksum_sha256
+        if object_present is True
+        and effect.expected_storage_checksum_sha256 is not None
+        else False if object_present is False else None
+    )
+    verified = database_blob_present is not None and object_present is not None
+    return {
+        "verified": verified,
+        "checks": {
+            "database_reloaded": database_blob_present is not None,
+            "object_probed": object_present is not None,
+            "database_matches_request": database_matches,
+            "database_integrity_verified": database_integrity_verified,
+            "object_size_matches": object_size_matches,
+            "object_checksum_matches": object_checksum_matches,
+        },
+        "blob_id": effect.blob_id,
+        "database_blob_present": database_blob_present,
+        "database_matches_request": database_matches,
+        "database_integrity_verified": database_integrity_verified,
+        "object_present": object_present,
+        "observed_size_bytes": observed_size,
+        "observed_checksum_sha256": observed_checksum,
+        "expected_storage_size_bytes": effect.expected_storage_size_bytes,
+        "expected_storage_checksum_sha256": effect.expected_storage_checksum_sha256,
+    }
+
+
+def _blob_write_complete(evidence: dict[str, object]) -> bool:
+    return bool(
+        evidence.get("verified") is True
+        and evidence.get("database_matches_request") is True
+        and evidence.get("database_integrity_verified") is True
+        and _object_matches(evidence)
+    )
+
+
+def _object_matches(evidence: dict[str, object]) -> bool:
+    checks = evidence.get("checks")
+    return bool(
+        isinstance(checks, dict)
+        and evidence.get("object_present") is True
+        and checks.get("object_size_matches") is True
+        and checks.get("object_checksum_matches") is True
+    )
+
+
+def _forward_complete_blob_repair(effect: PendingBlobWrite) -> bool:
+    try:
+        with effect.operation.session_factory() as session:
+            blob = session.get(FileBlob, effect.blob_id)
+            if (
+                blob is None
+                or blob.tenant_id != effect.tenant_id
+                or blob.storage_key != effect.storage_key
+                or blob.checksum_sha256 != effect.semantic_checksum_sha256
+                or blob.size_bytes != effect.semantic_size_bytes
+                or blob.protection_discriminator
+                != effect.protection_discriminator
+                or blob.encryption_envelope_id != effect.expected_envelope_id
+            ):
+                return False
+            blob.storage_checksum_sha256 = (
+                effect.expected_storage_checksum_sha256
+                if effect.expected_envelope_id
+                else None
+            )
+            blob.storage_size_bytes = (
+                effect.expected_storage_size_bytes
+                if effect.expected_envelope_id
+                else None
+            )
+            blob.integrity_status = "verified"
+            blob.integrity_checked_at = datetime.now(UTC)
+            blob.integrity_failure = None
+            blob.quarantined_at = None
+            session.add(blob)
+            session.commit()
+            return True
+    except Exception:
+        return False
+
+
+def _quarantine_blob_after_failed_verification(
+    effect: PendingBlobWrite,
+    evidence: dict[str, object],
+) -> None:
+    try:
+        with effect.operation.session_factory() as session:
+            blob = session.get(FileBlob, effect.blob_id)
+            if blob is None or blob.storage_key != effect.storage_key:
+                return
+            blob.integrity_status = (
+                "missing"
+                if evidence.get("object_present") is False
+                else "checksum_mismatch"
+            )
+            blob.integrity_checked_at = datetime.now(UTC)
+            blob.integrity_failure = "recovery_verification_failed"
+            blob.quarantined_at = datetime.now(UTC)
+            session.add(blob)
+            session.commit()
+    except Exception:
+        return
+
+
+def _orphan_cleanup_evidence(effect: PendingOrphanCleanup) -> dict[str, object]:
+    try:
+        object_present: bool | None = effect.backend.exists(effect.storage_key)
+    except (StorageBackendError, OSError):
+        object_present = None
+    finding_present: bool | None
+    finding_deleted: bool | None
+    still_unreferenced: bool | None
+    try:
+        with effect.operation.session_factory() as session:
+            finding = session.get(FileIntegrityFinding, effect.finding_id)
+            finding_present = finding is not None
+            finding_deleted = bool(
+                finding is not None and finding.state == "deleted"
+            ) if finding is not None else False
+            referenced = (
+                session.query(FileBlob.id)
+                .filter(
+                    FileBlob.tenant_id == effect.tenant_id,
+                    FileBlob.storage_key == effect.storage_key,
+                )
+                .first()
+            )
+            still_unreferenced = referenced is None
+    except Exception:
+        finding_present = None
+        finding_deleted = None
+        still_unreferenced = None
+    verified = object_present is not None and finding_present is not None
+    return {
+        "verified": verified,
+        "checks": {
+            "database_reloaded": finding_present is not None,
+            "object_probed": object_present is not None,
+            "finding_marked_deleted": finding_deleted,
+            "object_absent": (
+                not object_present if object_present is not None else None
+            ),
+            "still_unreferenced": still_unreferenced,
+        },
+        "finding_id": effect.finding_id,
+        "finding_present": finding_present,
+        "finding_deleted": finding_deleted,
+        "object_present": object_present,
+        "still_unreferenced": still_unreferenced,
+    }
+
+
+def _orphan_cleanup_complete(evidence: dict[str, object]) -> bool:
+    return bool(
+        evidence.get("verified") is True
+        and evidence.get("finding_deleted") is True
+        and evidence.get("object_present") is False
+        and evidence.get("still_unreferenced") is True
+    )
+
+
+def _forward_complete_orphan_finding(effect: PendingOrphanCleanup) -> bool:
+    try:
+        with effect.operation.session_factory() as session:
+            finding = session.get(FileIntegrityFinding, effect.finding_id)
+            if (
+                finding is None
+                or finding.tenant_id != effect.tenant_id
+                or finding.storage_key != effect.storage_key
+            ):
+                return False
+            scan = session.get(FileIntegrityScan, finding.scan_id)
+            if scan is None or not finding.storage_key.startswith(
+                scan.storage_prefix
+            ):
+                return False
+            referenced = (
+                session.query(FileBlob.id)
+                .filter(
+                    FileBlob.tenant_id == effect.tenant_id,
+                    FileBlob.storage_key == effect.storage_key,
+                )
+                .first()
+            )
+            if referenced:
+                return False
+            finding.state = "deleted"
+            finding.resolved_at = datetime.now(UTC)
+            finding.resolved_by_user_id = effect.resolved_by_user_id
+            session.add(finding)
+            session.commit()
+            return True
+    except Exception:
+        return False
+
+
+__all__ = [
+    "PendingBlobWrite",
+    "PendingOrphanCleanup",
+    "begin_blob_write_recovery",
+    "begin_orphan_cleanup_recovery",
+]

src/govoplan_files/backend/storage/sdk_peer_pinning.py

@@ -0,0 +1,331 @@
+from __future__ import annotations
+
+import inspect
+import socket
+import threading
+from functools import lru_cache
+from importlib import import_module
+from typing import Any
+
+from govoplan_core.security.outbound_http import (
+    OutboundHttpError,
+    create_outbound_connection,
+)
+
+
+class SdkPeerPinningError(RuntimeError):
+    """Raised when an optional SDK cannot be bound to the pinned transport."""
+
+
+_BOTOCORE_CLIENT_CREATION_LOCK = threading.Lock()
+
+
+def create_pinned_s3_client(**kwargs: Any) -> Any:
+    """Construct an S3 client whose first and every later socket is pinned.
+
+    Botocore does not expose the HTTP-session class through boto3's public
+    client API. Its endpoint creator does expose that seam, so this function
+    replaces the creator only for the bounded client-construction operation.
+    The lock avoids an unsafe interleaving with another GovOPlaN client
+    construction. Any unrelated botocore client created during the short
+    replacement window also receives the stricter transport.
+    """
+
+    try:
+        boto3 = import_module("boto3")
+        botocore_args = import_module("botocore.args")
+    except ImportError as exc:
+        raise SdkPeerPinningError(
+            "S3 connector browsing requires the optional boto3 dependency"
+        ) from exc
+
+    pinned_session_cls = _botocore_transport_types()[0]
+    original_creator = getattr(botocore_args, "EndpointCreator", None)
+    if original_creator is None or not hasattr(original_creator, "create_endpoint"):
+        raise SdkPeerPinningError(
+            "The installed botocore release does not expose the endpoint transport seam required for peer pinning"
+        )
+
+    class PinnedEndpointCreator(original_creator):  # type: ignore[misc, valid-type]
+        def create_endpoint(self, *args: Any, **endpoint_kwargs: Any) -> Any:
+            endpoint_kwargs["http_session_cls"] = pinned_session_cls
+            return super().create_endpoint(*args, **endpoint_kwargs)
+
+    with _BOTOCORE_CLIENT_CREATION_LOCK:
+        current_creator = getattr(botocore_args, "EndpointCreator", None)
+        if current_creator is not original_creator:
+            raise SdkPeerPinningError(
+                "Botocore endpoint construction changed concurrently; refusing to create an unproven S3 client"
+            )
+        botocore_args.EndpointCreator = PinnedEndpointCreator
+        try:
+            session_type = getattr(getattr(boto3, "session", None), "Session", None)
+            if session_type is None:
+                raise SdkPeerPinningError(
+                    "The installed boto3 release does not expose its isolated session constructor"
+                )
+            client = session_type().client("s3", **kwargs)
+        finally:
+            if getattr(botocore_args, "EndpointCreator", None) is PinnedEndpointCreator:
+                botocore_args.EndpointCreator = original_creator
+
+    http_session = getattr(getattr(client, "_endpoint", None), "http_session", None)
+    if http_session is None or not isinstance(http_session, pinned_session_cls):
+        close = getattr(client, "close", None)
+        if callable(close):
+            close()
+        raise SdkPeerPinningError(
+            "Botocore did not install the required pinned HTTP transport; the S3 client was discarded"
+        )
+    return client
+
+
+@lru_cache(maxsize=1)
+def _botocore_transport_types() -> tuple[type[Any], type[Any], type[Any]]:
+    try:
+        awsrequest = import_module("botocore.awsrequest")
+        httpsession = import_module("botocore.httpsession")
+        urllib3_exceptions = import_module("urllib3.exceptions")
+    except ImportError as exc:
+        raise SdkPeerPinningError(
+            "S3 connector browsing requires a compatible botocore HTTP transport"
+        ) from exc
+
+    required = (
+        "AWSHTTPConnection",
+        "AWSHTTPSConnection",
+        "AWSHTTPConnectionPool",
+        "AWSHTTPSConnectionPool",
+    )
+    if any(not hasattr(awsrequest, name) for name in required) or not hasattr(
+        httpsession, "URLLib3Session"
+    ):
+        raise SdkPeerPinningError(
+            "The installed botocore release is missing the connection classes required for peer pinning"
+        )
+
+    def pinned_new_connection(connection: Any) -> socket.socket:
+        hostname = str(getattr(connection, "_dns_host", "") or "").strip()
+        port = int(getattr(connection, "port", 0) or 0)
+        if not hostname or not port:
+            raise urllib3_exceptions.NewConnectionError(
+                connection, "Pinned S3 connection is missing its target authority"
+            )
+        try:
+            return create_outbound_connection(
+                hostname,
+                port,
+                timeout=getattr(connection, "timeout", None),
+                source_address=getattr(connection, "source_address", None),
+                socket_options=getattr(connection, "socket_options", None),
+                label="S3 connector peer",
+            )
+        except socket.timeout as exc:
+            raise urllib3_exceptions.ConnectTimeoutError(
+                connection,
+                f"Connection to {hostname} timed out while selecting an approved peer",
+            ) from exc
+        except (OSError, OutboundHttpError, ValueError) as exc:
+            raise urllib3_exceptions.NewConnectionError(
+                connection,
+                f"S3 connector peer was rejected: {exc}",
+            ) from exc
+
+    class PinnedAWSHTTPConnection(awsrequest.AWSHTTPConnection):
+        _new_conn = pinned_new_connection
+
+    class PinnedAWSHTTPSConnection(awsrequest.AWSHTTPSConnection):
+        _new_conn = pinned_new_connection
+
+    class PinnedAWSHTTPConnectionPool(awsrequest.AWSHTTPConnectionPool):
+        ConnectionCls = PinnedAWSHTTPConnection
+
+    class PinnedAWSHTTPSConnectionPool(awsrequest.AWSHTTPSConnectionPool):
+        ConnectionCls = PinnedAWSHTTPSConnection
+
+    class PinnedURLLib3Session(httpsession.URLLib3Session):
+        def __init__(self, *args: Any, **kwargs: Any) -> None:
+            # A proxy would select the final target outside this process. Until
+            # proxy peer delegation is modeled, connector traffic is direct.
+            kwargs["proxies"] = {}
+            super().__init__(*args, **kwargs)
+            self._pool_classes_by_scheme = {
+                "http": PinnedAWSHTTPConnectionPool,
+                "https": PinnedAWSHTTPSConnectionPool,
+            }
+            manager = getattr(self, "_manager", None)
+            if manager is None or not hasattr(manager, "pool_classes_by_scheme"):
+                raise SdkPeerPinningError(
+                    "The installed botocore pool manager cannot enforce pinned connection classes"
+                )
+            manager.pool_classes_by_scheme = self._pool_classes_by_scheme
+
+    return PinnedURLLib3Session, PinnedAWSHTTPConnection, PinnedAWSHTTPSConnection
+
+
+@lru_cache(maxsize=1)
+def pinned_smb_connection_cache() -> dict[str, Any]:
+    """Return the Files-owned cache; unproven process-global sessions are never reused."""
+
+    return {}
+
+
+def install_pinned_smb_transport(smbclient: Any) -> Any:
+    """Install a process-wide fail-closed smbclient session factory.
+
+    smbclient routes initial connections, reconnects, and DFS targets through
+    ``smbclient._pool.register_session``. Replacing that single factory with a
+    behavior-compatible implementation ensures every target uses a socket
+    selected by Core's deployment-wide outbound policy.
+    """
+
+    try:
+        pool = import_module("smbclient._pool")
+        connection_module = import_module("smbprotocol.connection")
+        session_module = import_module("smbprotocol.session")
+        transport_module = import_module("smbprotocol.transport")
+    except ImportError as exc:
+        raise SdkPeerPinningError(
+            "SMB connector browsing requires the optional smbprotocol dependency"
+        ) from exc
+
+    current = getattr(pool, "register_session", None)
+    if getattr(current, "__govoplan_peer_pinned__", False):
+        expected_tcp = getattr(current, "__govoplan_pinned_tcp__", None)
+        if expected_tcp is None or getattr(connection_module, "Tcp", None) is not expected_tcp:
+            raise SdkPeerPinningError(
+                "The installed SMB transport changed after peer pinning; refusing to reuse the session factory"
+            )
+        return smbclient
+    required_parameters = {
+        "server",
+        "username",
+        "password",
+        "port",
+        "encrypt",
+        "connection_timeout",
+        "connection_cache",
+        "auth_protocol",
+        "require_signing",
+    }
+    if current is None or not required_parameters.issubset(
+        inspect.signature(current).parameters
+    ):
+        raise SdkPeerPinningError(
+            "The installed smbprotocol release does not expose the session seam required for peer pinning"
+        )
+
+    class PinnedTcp(transport_module.Tcp):
+        def connect(self) -> None:
+            with self._sock_lock:
+                if self.connected:
+                    return
+                try:
+                    self._sock = create_outbound_connection(
+                        self.server,
+                        int(self.port),
+                        timeout=self.timeout,
+                        label="SMB connector peer",
+                    )
+                except (OSError, OutboundHttpError, ValueError) as exc:
+                    raise ValueError(
+                        f"SMB connector peer '{self.server}:{self.port}' was rejected: {exc}"
+                    ) from exc
+                self._sock.settimeout(None)
+                self.connected = True
+
+    # Connection.connect() instantiates the module-level Tcp symbol on every
+    # reconnect. Replacing it process-wide is deliberate: an SMB connection
+    # created by another code path must become stricter, never bypass Files'
+    # peer boundary.
+    if not hasattr(connection_module, "Tcp"):
+        raise SdkPeerPinningError(
+            "The installed smbprotocol release cannot install the pinned TCP transport"
+        )
+    connection_module.Tcp = PinnedTcp
+
+    def pinned_register_session(
+        server: str,
+        username: str | None = None,
+        password: str | None = None,
+        port: int = 445,
+        encrypt: bool | None = None,
+        connection_timeout: float = 60,
+        connection_cache: dict[str, Any] | None = None,
+        auth_protocol: str = "negotiate",
+        require_signing: bool = True,
+    ) -> Any:
+        cache = pinned_smb_connection_cache() if connection_cache is None else connection_cache
+        connection_key = f"{server.lower()}:{port}"
+        connection = cache.get(connection_key)
+        transport = getattr(connection, "transport", None)
+        if connection is not None and not isinstance(transport, PinnedTcp):
+            disconnect = getattr(connection, "disconnect", None)
+            if callable(disconnect):
+                try:
+                    disconnect(close=True)
+                except Exception:
+                    pass
+            cache.pop(connection_key, None)
+            connection = None
+        if connection is None or not getattr(connection.transport, "connected", False):
+            connection = connection_module.Connection(
+                pool.ClientConfig().client_guid,
+                server,
+                port,
+                require_signing=require_signing,
+            )
+            connection.transport = PinnedTcp(server, port)
+            connection.connect(timeout=connection_timeout)
+            if not isinstance(connection.transport, PinnedTcp):
+                disconnect = getattr(connection, "disconnect", None)
+                if callable(disconnect):
+                    try:
+                        disconnect(close=True)
+                    except Exception:
+                        pass
+                raise SdkPeerPinningError(
+                    "smbprotocol replaced the required pinned TCP transport during connection setup"
+                )
+            cache[connection_key] = connection
+
+        session = next(
+            (
+                item
+                for item in connection.session_table.values()
+                if username is None or item.username == username
+            ),
+            None,
+        )
+        if session is None:
+            session = session_module.Session(
+                connection,
+                username=username,
+                password=password,
+                require_encryption=(encrypt is True),
+                auth_protocol=auth_protocol,
+            )
+            session.connect()
+        elif encrypt is not None:
+            if session.encrypt_data and not encrypt:
+                raise ValueError(
+                    "Cannot disable encryption on an already negotiated session."
+                )
+            if not session.encrypt_data and encrypt:
+                session.encrypt = True
+        return session
+
+    pinned_register_session.__govoplan_peer_pinned__ = True  # type: ignore[attr-defined]
+    pinned_register_session.__govoplan_pinned_tcp__ = PinnedTcp  # type: ignore[attr-defined]
+    pool.register_session = pinned_register_session
+    if hasattr(smbclient, "register_session"):
+        smbclient.register_session = pinned_register_session
+    return smbclient
+
+
+__all__ = [
+    "SdkPeerPinningError",
+    "create_pinned_s3_client",
+    "install_pinned_smb_transport",
+    "pinned_smb_connection_cache",
+]

src/govoplan_files/backend/storage/share_state.py

@@ -0,0 +1,34 @@
+from __future__ import annotations
+
+from datetime import datetime, timezone
+
+from sqlalchemy import and_, or_
+
+from govoplan_files.backend.db.models import FileShare
+from govoplan_files.backend.storage.common import utcnow
+
+
+def effective_file_share_clause(*, at: datetime | None = None):
+    checked_at = at or utcnow()
+    return and_(
+        FileShare.revoked_at.is_(None),
+        or_(FileShare.expires_at.is_(None), FileShare.expires_at > checked_at),
+    )
+
+
+def file_share_is_active(share: FileShare, *, at: datetime | None = None) -> bool:
+    if share.revoked_at is not None:
+        return False
+    if share.expires_at is None:
+        return True
+    checked_at = _aware_utc(at or utcnow())
+    return _aware_utc(share.expires_at) > checked_at
+
+
+def _aware_utc(value: datetime) -> datetime:
+    if value.tzinfo is None:
+        return value.replace(tzinfo=timezone.utc)
+    return value.astimezone(timezone.utc)
+
+
+__all__ = ["effective_file_share_clause", "file_share_is_active"]

tests/test_access_provider.py

@@ -1,15 +1,23 @@
 from __future__ import annotations
 
 import unittest
+from datetime import datetime, timedelta, timezone
+from unittest.mock import patch
 
-from sqlalchemy import create_engine
+from sqlalchemy import create_engine, text
 from sqlalchemy.orm import sessionmaker
 
 from govoplan_access.backend.db.models import Account, Group, User
 from govoplan_core.core.access import PrincipalRef
+from govoplan_core.core.change_sequence import ChangeSequenceEntry
 from govoplan_core.db.base import Base
 from govoplan_files.backend.capabilities import FilesAccessService, virtual_folder_resource_id
 from govoplan_files.backend.db.models import FileAsset, FileFolder, FileShare
+from govoplan_files.backend.storage.files import (
+    count_assets_for_user,
+    list_assets_for_user,
+    list_recent_assets_for_user,
+)
 
 
 TENANT_ID = "tenant-1"
@@ -19,6 +27,54 @@ GROUP_ID = "group-1"
 
 
 class FilesAccessProviderTests(unittest.TestCase):
+    def test_recent_projection_is_bounded_ordered_and_access_filtered(self) -> None:
+        session = _session()
+        self.addCleanup(_close_session, session)
+        _seed_access_subjects(session)
+        now = datetime.now(timezone.utc)
+        session.add_all(
+            [
+                FileAsset(
+                    id="owned-older",
+                    tenant_id=TENANT_ID,
+                    owner_type="user",
+                    owner_user_id=USER_ID,
+                    display_path="older.pdf",
+                    filename="older.pdf",
+                    updated_at=now - timedelta(hours=2),
+                ),
+                FileAsset(
+                    id="owned-newer",
+                    tenant_id=TENANT_ID,
+                    owner_type="user",
+                    owner_user_id=USER_ID,
+                    display_path="newer.pdf",
+                    filename="newer.pdf",
+                    updated_at=now - timedelta(hours=1),
+                ),
+                FileAsset(
+                    id="other-newest",
+                    tenant_id=TENANT_ID,
+                    owner_type="user",
+                    owner_user_id=OTHER_USER_ID,
+                    display_path="private.pdf",
+                    filename="private.pdf",
+                    updated_at=now,
+                ),
+            ]
+        )
+        session.commit()
+
+        with patch("govoplan_files.backend.storage.files.user_group_ids", return_value=[]):
+            recent = list_recent_assets_for_user(
+                session,
+                tenant_id=TENANT_ID,
+                user_id=USER_ID,
+                limit=1,
+            )
+
+        self.assertEqual(["owned-newer"], [asset.id for asset in recent])
+
     def test_file_access_provider_explains_owner_share_admin_and_missing_resources(self) -> None:
         session = _session()
         self.addCleanup(_close_session, session)
@@ -75,10 +131,123 @@ class FilesAccessProviderTests(unittest.TestCase):
         self.assertTrue(any(item.kind == "owner" and item.id == GROUP_ID for item in virtual_items))
         self.assertEqual("files.not_found", missing_items[0].source)
 
+    def test_file_property_filters_cover_campaign_and_audit_usage(self) -> None:
+        session = _session()
+        self.addCleanup(_close_session, session)
+        _seed_access_subjects(session)
+        assets = [
+            FileAsset(
+                id=f"file-{index}",
+                tenant_id=TENANT_ID,
+                owner_type="user",
+                owner_user_id=USER_ID,
+                display_path=f"{index}.pdf",
+                filename=f"{index}.pdf",
+            )
+            for index in range(1, 4)
+        ]
+        session.add_all([
+            *assets,
+            FileShare(
+                id="share-campaign",
+                tenant_id=TENANT_ID,
+                file_asset_id=assets[0].id,
+                target_type="campaign",
+                target_id="campaign-1",
+                permission="read",
+            ),
+            FileShare(
+                id="share-campaign-expired",
+                tenant_id=TENANT_ID,
+                file_asset_id=assets[2].id,
+                target_type="campaign",
+                target_id="campaign-1",
+                permission="read",
+                expires_at=datetime.now(timezone.utc) - timedelta(seconds=1),
+            ),
+        ])
+        session.commit()
+        session.execute(
+            text(
+                "INSERT INTO campaign_attachment_uses "
+                "(id, tenant_id, file_asset_id, use_stage) "
+                "VALUES (:id, :tenant_id, :file_asset_id, :use_stage)"
+            ),
+            {
+                "id": "attachment-use-1",
+                "tenant_id": TENANT_ID,
+                "file_asset_id": assets[1].id,
+                "use_stage": "sent",
+            },
+        )
+        session.commit()
+
+        linked = list_assets_for_user(
+            session,
+            tenant_id=TENANT_ID,
+            user_id=USER_ID,
+            owner_type="user",
+            owner_id=USER_ID,
+            campaign_usage="linked",
+        )
+        unlinked = list_assets_for_user(
+            session,
+            tenant_id=TENANT_ID,
+            user_id=USER_ID,
+            owner_type="user",
+            owner_id=USER_ID,
+            campaign_usage="unlinked",
+        )
+        audit_relevant = list_assets_for_user(
+            session,
+            tenant_id=TENANT_ID,
+            user_id=USER_ID,
+            owner_type="user",
+            owner_id=USER_ID,
+            audit_relevant=True,
+        )
+
+        self.assertEqual([asset.id for asset in linked], ["file-1", "file-2"])
+        self.assertEqual([asset.id for asset in unlinked], ["file-3"])
+        self.assertEqual([asset.id for asset in audit_relevant], ["file-2"])
+        self.assertEqual(
+            count_assets_for_user(
+                session,
+                tenant_id=TENANT_ID,
+                user_id=USER_ID,
+                owner_type="user",
+                owner_id=USER_ID,
+                campaign_usage="unlinked",
+            ),
+            1,
+        )
+
 
 def _session():
     engine = create_engine("sqlite:///:memory:", future=True)
-    Base.metadata.create_all(bind=engine, tables=[Account.__table__, User.__table__, Group.__table__, FileAsset.__table__, FileFolder.__table__, FileShare.__table__])
+    Base.metadata.create_all(
+        bind=engine,
+        tables=[
+            Account.__table__,
+            User.__table__,
+            Group.__table__,
+            ChangeSequenceEntry.__table__,
+            FileAsset.__table__,
+            FileFolder.__table__,
+            FileShare.__table__,
+        ],
+    )
+    with engine.begin() as connection:
+        connection.execute(
+            text(
+                "CREATE TABLE campaign_attachment_uses ("
+                "id VARCHAR(36) PRIMARY KEY, "
+                "tenant_id VARCHAR(36) NOT NULL, "
+                "file_asset_id VARCHAR(36) NOT NULL, "
+                "use_stage VARCHAR(20) NOT NULL"
+                ")"
+            )
+        )
     return sessionmaker(bind=engine, future=True)()
 
 

tests/test_archive_preview_token.py

@@ -0,0 +1,101 @@
+from __future__ import annotations
+
+import io
+import hashlib
+import unittest
+import zipfile
+from types import SimpleNamespace
+from unittest.mock import patch
+
+from fastapi import UploadFile
+
+from govoplan_files.backend.routes.uploads import (
+    _validate_archive_preview_token,
+    preview_archive_upload,
+)
+from govoplan_files.backend.storage.common import FileStorageError
+
+
+def _archive_bytes() -> bytes:
+    target = io.BytesIO()
+    with zipfile.ZipFile(target, "w", compression=zipfile.ZIP_DEFLATED) as archive:
+        archive.writestr("folder/report.txt", b"report")
+    return target.getvalue()
+
+
+class ArchivePreviewTokenTests(unittest.TestCase):
+    def setUp(self) -> None:
+        self.settings = SimpleNamespace(
+            file_upload_zip_max_bytes=10 * 1024 * 1024,
+            file_archive_max_entries=10_000,
+            file_archive_max_expanded_bytes=2 * 1024 * 1024 * 1024,
+            file_archive_max_expansion_ratio=100,
+            file_archive_preview_ttl_seconds=30 * 60,
+        )
+        self.principal = SimpleNamespace(
+            tenant_id="tenant-1",
+            user=SimpleNamespace(id="user-1"),
+        )
+
+    def test_preview_token_is_bound_to_actor_archive_and_destination(self) -> None:
+        archive_data = _archive_bytes()
+        upload = UploadFile(
+            file=io.BytesIO(archive_data),
+            filename="monthly.zip",
+        )
+        with patch(
+            "govoplan_files.backend.routes.uploads.settings",
+            self.settings,
+        ):
+            preview = preview_archive_upload(
+                file=upload,
+                owner_type="user",
+                owner_id="user-1",
+                path="incoming",
+                campaign_id=None,
+                password=None,
+                principal=self.principal,
+            )
+            _validate_archive_preview_token(
+                preview.preview_token,
+                tenant_id="tenant-1",
+                user_id="user-1",
+                owner_type="user",
+                owner_id="user-1",
+                path="incoming",
+                campaign_id=None,
+                archive_format="zip",
+                archive_sha256=hashlib.sha256(archive_data).hexdigest(),
+            )
+            with self.assertRaisesRegex(
+                FileStorageError, "does not match this upload destination"
+            ):
+                _validate_archive_preview_token(
+                    preview.preview_token,
+                    tenant_id="tenant-1",
+                    user_id="user-1",
+                    owner_type="user",
+                    owner_id="user-1",
+                    path="elsewhere",
+                    campaign_id=None,
+                    archive_format="zip",
+                    archive_sha256=hashlib.sha256(archive_data).hexdigest(),
+                )
+            with self.assertRaisesRegex(
+                FileStorageError, "contents changed"
+            ):
+                _validate_archive_preview_token(
+                    preview.preview_token,
+                    tenant_id="tenant-1",
+                    user_id="user-1",
+                    owner_type="user",
+                    owner_id="user-1",
+                    path="incoming",
+                    campaign_id=None,
+                    archive_format="zip",
+                    archive_sha256="0" * 64,
+                )
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_archives.py

@@ -0,0 +1,170 @@
+from __future__ import annotations
+
+import io
+import tarfile
+import unittest
+import zipfile
+from types import SimpleNamespace
+from unittest.mock import patch
+
+import pyzipper
+
+from govoplan_files.backend.storage.archives import (
+    ArchivePasswordError,
+    extract_archive_upload,
+    inspect_archive,
+)
+from govoplan_files.backend.storage.common import FileStorageError
+
+
+def _zip_bytes(entries: dict[str, bytes]) -> bytes:
+    target = io.BytesIO()
+    with zipfile.ZipFile(target, "w", compression=zipfile.ZIP_DEFLATED) as archive:
+        for path, data in entries.items():
+            archive.writestr(path, data)
+    return target.getvalue()
+
+
+def _tar_bytes(entries: dict[str, bytes], mode: str = "w:gz") -> bytes:
+    target = io.BytesIO()
+    with tarfile.open(fileobj=target, mode=mode) as archive:
+        for path, data in entries.items():
+            info = tarfile.TarInfo(path)
+            info.size = len(data)
+            archive.addfile(info, io.BytesIO(data))
+    return target.getvalue()
+
+
+def _encrypted_zip_bytes(password: str) -> bytes:
+    target = io.BytesIO()
+    with pyzipper.AESZipFile(
+        target,
+        "w",
+        compression=zipfile.ZIP_DEFLATED,
+        encryption=pyzipper.WZ_AES,
+    ) as archive:
+        archive.setpassword(password.encode("utf-8"))
+        archive.writestr("secure/report.txt", b"classified")
+    return target.getvalue()
+
+
+class ArchiveInspectionTests(unittest.TestCase):
+    def test_zip_preview_derives_folders_and_sizes(self) -> None:
+        inspection = inspect_archive(
+            _zip_bytes(
+                {
+                    "department/one.txt": b"one",
+                    "department/nested/two.csv": b"two",
+                }
+            ),
+            filename="monthly.zip",
+        )
+
+        self.assertEqual("zip", inspection.archive_format)
+        self.assertEqual(2, inspection.file_count)
+        self.assertEqual(2, inspection.directory_count)
+        self.assertEqual(6, inspection.expanded_size_bytes)
+        self.assertEqual(
+            [
+                "department",
+                "department/nested",
+                "department/nested/two.csv",
+                "department/one.txt",
+            ],
+            [entry.path for entry in inspection.entries],
+        )
+
+    def test_tar_compression_variants_are_inspected(self) -> None:
+        for filename, mode in (
+            ("monthly.tar", "w"),
+            ("monthly.tar.gz", "w:gz"),
+            ("monthly.tar.bz2", "w:bz2"),
+            ("monthly.tar.xz", "w:xz"),
+        ):
+            with self.subTest(filename=filename):
+                inspection = inspect_archive(
+                    _tar_bytes({"report.txt": b"report"}, mode=mode),
+                    filename=filename,
+                )
+                self.assertEqual(1, inspection.file_count)
+                self.assertEqual(filename.removeprefix("monthly."), inspection.archive_format)
+
+    def test_unsafe_member_path_is_rejected(self) -> None:
+        with self.assertRaisesRegex(FileStorageError, "Unsafe archive member"):
+            inspect_archive(
+                _zip_bytes({"../escape.txt": b"no"}),
+                filename="unsafe.zip",
+            )
+
+    def test_archive_bomb_ratio_is_rejected(self) -> None:
+        with self.assertRaisesRegex(FileStorageError, "expansion ratio"):
+            inspect_archive(
+                _zip_bytes({"zeros.bin": b"\0" * 50_000}),
+                filename="bomb.zip",
+                max_expansion_ratio=2,
+            )
+
+    def test_encrypted_zip_reports_and_verifies_password(self) -> None:
+        archive = _encrypted_zip_bytes("correct horse")
+
+        missing = inspect_archive(archive, filename="secure.zip")
+        self.assertTrue(missing.requires_password)
+        self.assertFalse(missing.password_verified)
+
+        verified = inspect_archive(
+            archive,
+            filename="secure.zip",
+            password="correct horse",
+        )
+        self.assertTrue(verified.requires_password)
+        self.assertTrue(verified.password_verified)
+
+        with self.assertRaisesRegex(ArchivePasswordError, "incorrect"):
+            inspect_archive(
+                archive,
+                filename="secure.zip",
+                password="wrong",
+            )
+
+    def test_folder_selection_extracts_only_descendants(self) -> None:
+        archive = _zip_bytes(
+            {
+                "selected/one.txt": b"one",
+                "selected/two.txt": b"two",
+                "other/three.txt": b"three",
+            }
+        )
+        stored = SimpleNamespace(
+            asset=SimpleNamespace(id="asset"),
+            version=SimpleNamespace(id="version"),
+            blob=SimpleNamespace(id="blob"),
+        )
+        with patch(
+            "govoplan_files.backend.storage.archives.create_file_asset",
+            return_value=stored,
+        ) as create_asset:
+            result = extract_archive_upload(
+                object(),  # type: ignore[arg-type]
+                tenant_id="tenant-1",
+                owner_type="user",
+                owner_id="user-1",
+                user_id="user-1",
+                archive_data=archive,
+                filename="selection.zip",
+                folder="imports",
+                campaign_id=None,
+                selected_paths=["selected"],
+            )
+
+        self.assertEqual(2, len(result))
+        self.assertEqual(
+            {"imports/selected/one.txt", "imports/selected/two.txt"},
+            {
+                call.kwargs["display_path"]
+                for call in create_asset.call_args_list
+            },
+        )
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_artifact_store.py

@@ -0,0 +1,89 @@
+from __future__ import annotations
+
+from types import SimpleNamespace
+import unittest
+from unittest.mock import patch
+
+from govoplan_core.auth import ApiPrincipal
+from govoplan_core.core.access import PrincipalRef
+from govoplan_core.core.files import ManagedArtifactStore, ManagedArtifactWriteRequest
+from govoplan_files.backend.capabilities import FilesArtifactStore
+
+
+def principal() -> ApiPrincipal:
+    return ApiPrincipal(
+        principal=PrincipalRef(
+            account_id="account-1",
+            membership_id="membership-1",
+            tenant_id="tenant-1",
+            scopes=frozenset({"files:file:upload"}),
+        ),
+        account=object(),
+        user=SimpleNamespace(id="user-1"),
+    )
+
+
+class _Session:
+    def query(self):
+        raise AssertionError("not used")
+
+    def flush(self):
+        raise AssertionError("not used")
+
+
+def stored():
+    return SimpleNamespace(
+        asset=SimpleNamespace(
+            id="file-1",
+            display_path="Generated/Templates/result.html",
+            owner_type="user",
+        ),
+        version=SimpleNamespace(
+            id="version-1",
+            filename_at_upload="result.html",
+            content_type="text/html",
+            size_bytes=6,
+            checksum_sha256="0" * 64,
+        ),
+    )
+
+
+class FilesArtifactStoreTests(unittest.TestCase):
+    def test_plain_write_uses_files_authority_and_returns_provider_neutral_ref(self) -> None:
+        service = FilesArtifactStore()
+        self.assertIsInstance(service, ManagedArtifactStore)
+        request = ManagedArtifactWriteRequest(
+            filename="result.html",
+            payload=b"result",
+            content_type="text/html",
+            folder="Generated/Templates",
+            metadata={"producer_module": "templates"},
+        )
+        with patch("govoplan_files.backend.capabilities.create_file_asset", return_value=stored()) as create:
+            result = service.store_artifact(_Session(), principal(), request=request)
+        self.assertEqual("file-1", result.file_asset_id)
+        self.assertEqual("version-1", result.file_version_id)
+        self.assertEqual("tenant-1", create.call_args.kwargs["tenant_id"])
+        self.assertEqual("user-1", create.call_args.kwargs["owner_id"])
+        self.assertEqual(b"result", create.call_args.kwargs["data"])
+
+    def test_idempotent_write_uses_source_provenance_without_payload_metadata(self) -> None:
+        request = ManagedArtifactWriteRequest(
+            filename="result.html",
+            payload=b"secret body",
+            content_type="text/html",
+            idempotency_key="render-1",
+            metadata={"producer_module": "templates", "output_sha256": "a" * 64},
+        )
+        with patch(
+            "govoplan_files.backend.capabilities.sync_file_asset_from_source",
+            return_value=(stored(), "unchanged", None),
+        ) as sync:
+            FilesArtifactStore().store_artifact(_Session(), principal(), request=request)
+        metadata = sync.call_args.kwargs["metadata"]
+        self.assertEqual("render-1", metadata["source_provenance"]["external_id"])
+        self.assertNotIn("secret body", str(metadata))
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_configuration_provider.py

@@ -0,0 +1,219 @@
+from __future__ import annotations
+
+from types import SimpleNamespace
+import unittest
+
+from govoplan_core.core.configuration_packages import (
+    ConfigurationPackageFragment,
+    ConfigurationPreflightContext,
+)
+from govoplan_core.core.infrastructure_capabilities import (
+    infrastructure_capability_receipt_from_mapping,
+)
+from govoplan_files.backend.configuration_provider import (
+    FILES_CONFIGURATION_CAPABILITY,
+    FilesConfigurationProvider,
+)
+from govoplan_files.backend.manifest import manifest
+
+
+def _receipt(*, source: str = "host-local", state: str = "configured"):
+    endpoint = (
+        {"kind": "filesystem", "reference": "volume:files-data"}
+        if source == "host-local"
+        else {"scheme": "http", "host": "garage", "port": 3900}
+    )
+    secret_refs = (
+        []
+        if source == "host-local"
+        else [
+            "env:FILE_STORAGE_S3_ACCESS_KEY_ID",
+            "env:FILE_STORAGE_S3_SECRET_ACCESS_KEY",
+            "env:GARAGE_RPC_SECRET",
+        ]
+    )
+    return infrastructure_capability_receipt_from_mapping(
+        {
+            "schema_version": 1,
+            "installation_id": "files-provider-test",
+            "profile": "evaluation",
+            "capabilities": [
+                {
+                    "id": "files.storage",
+                    "label": "Managed file content storage",
+                    "state": state,
+                    "source": source,
+                    "detail": "Deployment-owned storage binding.",
+                    "endpoint": endpoint,
+                    "secret_refs": secret_refs,
+                    "dependent_modules": ["files"],
+                }
+            ],
+            "post_install_tasks": [],
+        }
+    )
+
+
+def _local_settings(**overrides):
+    values = {
+        "file_storage_backend": "local",
+        "file_storage_local_root": "/var/lib/govoplan/files",
+    }
+    values.update(overrides)
+    return SimpleNamespace(**values)
+
+
+def _s3_settings(**overrides):
+    values = {
+        "file_storage_backend": "s3",
+        "file_storage_s3_endpoint_url": "http://garage:3900",
+        "file_storage_s3_bucket": "files",
+        "file_storage_s3_deployment_managed": True,
+        "file_storage_s3_endpoint_trusted": False,
+    }
+    values.update(overrides)
+    return SimpleNamespace(**values)
+
+
+class FilesConfigurationProviderTests(unittest.TestCase):
+    def test_provider_is_registered(self) -> None:
+        self.assertIn(FILES_CONFIGURATION_CAPABILITY, manifest.capability_factories)
+
+    def test_matching_local_storage_is_an_idempotent_noop(self) -> None:
+        provider = FilesConfigurationProvider(
+            settings=_local_settings(),
+            environment={},
+        )
+        context = ConfigurationPreflightContext(
+            infrastructure_receipt=_receipt(),
+        )
+        fragment = ConfigurationPackageFragment(
+            module_id="files",
+            fragment_type="managed_storage",
+            fragment_id="files-storage",
+            payload={
+                "expected_backend": "local",
+                "expected_source": "host-local",
+            },
+        )
+
+        first = provider.preflight(fragment, context)
+        applied = provider.apply(fragment, {}, context)
+        second = provider.preflight(fragment, context)
+
+        self.assertEqual("skip", first.plan[0].action)
+        self.assertEqual("skip", second.plan[0].action)
+        self.assertEqual((), applied.diagnostics)
+        self.assertEqual({}, applied.created_refs)
+        self.assertEqual({}, applied.updated_refs)
+
+    def test_runtime_backend_mismatch_blocks_without_rewriting_settings(self) -> None:
+        settings = _local_settings()
+        provider = FilesConfigurationProvider(settings=settings, environment={})
+        context = ConfigurationPreflightContext(
+            infrastructure_receipt=_receipt(source="installer-managed-garage"),
+        )
+        fragment = ConfigurationPackageFragment(
+            module_id="files",
+            fragment_type="managed_storage",
+            payload={},
+        )
+
+        result = provider.preflight(fragment, context)
+
+        self.assertEqual("blocked", result.plan[0].action)
+        self.assertIn(
+            "files_storage_runtime_mismatch",
+            {item.code for item in result.diagnostics},
+        )
+        self.assertEqual("local", settings.file_storage_backend)
+
+    def test_garage_binding_requires_only_files_secret_references(self) -> None:
+        fragment = ConfigurationPackageFragment(
+            module_id="files",
+            fragment_type="managed_storage",
+            payload={},
+        )
+        context = ConfigurationPreflightContext(
+            infrastructure_receipt=_receipt(source="installer-managed-garage"),
+        )
+        missing = FilesConfigurationProvider(
+            settings=_s3_settings(),
+            environment={},
+        ).preflight(fragment, context)
+        available = FilesConfigurationProvider(
+            settings=_s3_settings(),
+            environment={
+                "FILE_STORAGE_S3_ACCESS_KEY_ID": "reference-resolved",
+                "FILE_STORAGE_S3_SECRET_ACCESS_KEY": "reference-resolved",
+            },
+        ).preflight(fragment, context)
+
+        self.assertEqual("blocked", missing.plan[0].action)
+        self.assertEqual(
+            2,
+            sum(
+                item.code == "files_storage_secret_reference_unresolved"
+                for item in missing.diagnostics
+            ),
+        )
+        self.assertEqual("skip", available.plan[0].action)
+
+    def test_operator_supplied_s3_requires_explicit_endpoint_trust(self) -> None:
+        receipt = _receipt(source="operator-supplied-s3", state="externally_supplied")
+        fragment = ConfigurationPackageFragment(
+            module_id="files",
+            fragment_type="managed_storage",
+            payload={},
+        )
+        context = ConfigurationPreflightContext(infrastructure_receipt=receipt)
+        environment = {
+            "FILE_STORAGE_S3_ACCESS_KEY_ID": "reference-resolved",
+            "FILE_STORAGE_S3_SECRET_ACCESS_KEY": "reference-resolved",
+        }
+
+        blocked = FilesConfigurationProvider(
+            settings=_s3_settings(
+                file_storage_s3_deployment_managed=False,
+                file_storage_s3_endpoint_trusted=False,
+            ),
+            environment=environment,
+        ).preflight(fragment, context)
+        ready = FilesConfigurationProvider(
+            settings=_s3_settings(
+                file_storage_s3_deployment_managed=False,
+                file_storage_s3_endpoint_trusted=True,
+            ),
+            environment=environment,
+        ).preflight(fragment, context)
+
+        self.assertEqual("blocked", blocked.plan[0].action)
+        self.assertIn(
+            "files_storage_trust_boundary_missing",
+            {item.code for item in blocked.diagnostics},
+        )
+        self.assertEqual("skip", ready.plan[0].action)
+
+    def test_inline_storage_secret_is_rejected(self) -> None:
+        provider = FilesConfigurationProvider(
+            settings=_local_settings(),
+            environment={},
+        )
+        result = provider.preflight(
+            ConfigurationPackageFragment(
+                module_id="files",
+                fragment_type="managed_storage",
+                payload={"secret_access_key": "inline"},
+            ),
+            ConfigurationPreflightContext(infrastructure_receipt=_receipt()),
+        )
+
+        self.assertEqual("blocked", result.plan[0].action)
+        self.assertIn(
+            "files_configuration_secret_forbidden",
+            {item.code for item in result.diagnostics},
+        )
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_connector_credential_deletion.py

@@ -12,7 +12,8 @@ from govoplan_core.core.change_sequence import ChangeSequenceEntry
 from govoplan_core.db.base import Base
 from govoplan_core.security.secrets import encrypt_secret
 from govoplan_files.backend.db.models import FileConnectorCredential, FileConnectorProfile
-from govoplan_files.backend.router import deactivate_connector_credential, deactivate_connector_profile
+from govoplan_files.backend.routes.connector_profiles import deactivate_connector_profile
+from govoplan_files.backend.routes.connector_settings import deactivate_connector_credential
 
 
 class Principal:

tests/test_connector_deployment.py

@@ -10,7 +10,7 @@ from unittest.mock import MagicMock
 
 from fastapi import HTTPException
 
-from govoplan_files.backend.router import discover_connector_endpoint
+from govoplan_files.backend.routes.connector_settings import discover_connector_endpoint
 from govoplan_files.backend.schemas import FileConnectorDiscoveryRequest
 from govoplan_files.backend.storage.connector_browse import ConnectorBrowseError, _profile_password, _s3_verify
 from govoplan_files.backend.storage.connector_deployment import (
@@ -182,7 +182,7 @@ class ConnectorDiscoveryBoundaryTests(unittest.TestCase):
             credential_mode="basic",
             credentials={"username": "admin", "password_env": "MASTER_KEY_B64"},
         )
-        with patch("govoplan_files.backend.router.browse_connector_profile") as browse, self.assertRaises(
+        with patch("govoplan_files.backend.routes.connector_settings.browse_connector_profile") as browse, self.assertRaises(
             HTTPException
         ) as raised:
             discover_connector_endpoint(payload, session=object(), principal=self._principal())  # type: ignore[arg-type]
@@ -212,10 +212,10 @@ class ConnectorDiscoveryBoundaryTests(unittest.TestCase):
             events.append("io")
             return []
 
-        with patch("govoplan_files.backend.router._ensure_connector_configuration_allowed", side_effect=ensure), patch(
+        with patch("govoplan_files.backend.routes.connector_settings._ensure_connector_configuration_allowed", side_effect=ensure), patch(
-            "govoplan_files.backend.router._audit_connector_discovery_attempt",
+            "govoplan_files.backend.routes.connector_settings._audit_connector_discovery_attempt",
             side_effect=audit,
-        ), patch("govoplan_files.backend.router.browse_connector_profile", side_effect=browse):
+        ), patch("govoplan_files.backend.routes.connector_settings.browse_connector_profile", side_effect=browse):
             response = discover_connector_endpoint(payload, session=object(), principal=self._principal())  # type: ignore[arg-type]
 
         self.assertEqual("usable", response.status)

tests/test_connector_folder_sync.py

@@ -0,0 +1,303 @@
+from __future__ import annotations
+
+import unittest
+from datetime import UTC, datetime
+from types import SimpleNamespace
+from unittest.mock import MagicMock, Mock, call, patch
+
+from govoplan_files.backend.routes.connector_io import sync_connector_space_folder
+from govoplan_files.backend.schemas import (
+    FileAssetResponse,
+    FileConnectorFolderSyncRequest,
+)
+from govoplan_files.backend.storage.common import FileStorageError, UploadedStoredFile
+from govoplan_files.backend.storage.connector_browse import ConnectorBrowseItem
+from govoplan_files.backend.storage.connector_folder_sync import (
+    connector_relative_path,
+    discover_connector_folder,
+)
+from govoplan_files.backend.storage.connector_imports import ConnectorDownloadedFile
+from govoplan_files.backend.storage.connector_policy import ConnectorPolicyDenied
+from govoplan_files.backend.storage.connector_profiles import ConnectorProfile
+
+
+def _item(kind: str, path: str, *, token: str | None = None) -> ConnectorBrowseItem:
+    return ConnectorBrowseItem(
+        kind=kind,
+        name=path.rsplit("/", 1)[-1],
+        path=path,
+        external_id=f"external:{path}",
+        etag=f"etag:{path}",
+        metadata={"next_continuation_token": token} if token else {},
+    )
+
+
+class ConnectorFolderDiscoveryTests(unittest.TestCase):
+    def setUp(self) -> None:
+        self.profile = ConnectorProfile(
+            id="profile-1",
+            label="Documents",
+            provider="webdav",
+            capabilities=("browse", "sync"),
+        )
+
+    @patch("govoplan_files.backend.storage.connector_folder_sync.browse_connector_profile")
+    def test_recursively_discovers_files_across_pages_in_deterministic_order(
+        self, browse: Mock
+    ) -> None:
+        browse.side_effect = [
+            [_item("file", "root/z.txt", token="page-2")],
+            [_item("folder", "root/nested"), _item("file", "root/a.txt")],
+            [_item("file", "root/nested/b.txt")],
+        ]
+
+        result = discover_connector_folder(
+            self.profile,
+            path="root",
+            library_id="library-1",
+            recursive=True,
+            max_files=10,
+            max_depth=3,
+        )
+
+        self.assertEqual(
+            ["root/a.txt", "root/nested/b.txt", "root/z.txt"],
+            [item.path for item in result.files],
+        )
+        self.assertFalse(result.truncated)
+        self.assertEqual([], result.skipped)
+        self.assertEqual(
+            [
+                call(
+                    self.profile,
+                    path="root",
+                    library_id="library-1",
+                    continuation_token=None,
+                ),
+                call(
+                    self.profile,
+                    path="root",
+                    library_id="library-1",
+                    continuation_token="page-2",
+                ),
+                call(
+                    self.profile,
+                    path="root/nested",
+                    library_id="library-1",
+                    continuation_token=None,
+                ),
+            ],
+            browse.call_args_list,
+        )
+
+    @patch("govoplan_files.backend.storage.connector_folder_sync.browse_connector_profile")
+    def test_non_recursive_and_file_limit_skips_are_explicit(self, browse: Mock) -> None:
+        browse.return_value = [
+            _item("folder", "root/nested"),
+            _item("file", "root/a.txt"),
+            _item("file", "root/b.txt"),
+        ]
+
+        result = discover_connector_folder(
+            self.profile,
+            path="root",
+            library_id=None,
+            recursive=False,
+            max_files=1,
+            max_depth=0,
+        )
+
+        self.assertEqual(["root/a.txt"], [item.path for item in result.files])
+        self.assertTrue(result.truncated)
+        self.assertEqual(
+            ["root/nested", "root/b.txt"],
+            [item.path for item in result.skipped],
+        )
+
+    def test_relative_path_rejects_provider_results_outside_linked_root(self) -> None:
+        self.assertEqual(
+            "nested/report.csv",
+            connector_relative_path(
+                space_root="linked/root", item_path="linked/root/nested/report.csv"
+            ),
+        )
+        with self.assertRaisesRegex(ValueError, "outside"):
+            connector_relative_path(
+                space_root="linked/root", item_path="another/root/report.csv"
+            )
+
+
+class ConnectorFolderSyncRouteTests(unittest.TestCase):
+    def setUp(self) -> None:
+        self.session = MagicMock()
+        self.space = SimpleNamespace(
+            id="space-1",
+            tenant_id="tenant-1",
+            owner_type="user",
+            owner_user_id="user-1",
+            owner_group_id=None,
+            connector_profile_id="profile-1",
+            provider="webdav",
+            library_id="library-1",
+            remote_path="linked/root",
+            sync_mode="manual",
+        )
+        self.profile = ConnectorProfile(
+            id="profile-1",
+            label="Documents",
+            provider="webdav",
+            capabilities=("browse", "sync"),
+        )
+        self.principal = SimpleNamespace(
+            tenant_id="tenant-1", user=SimpleNamespace(id="user-1")
+        )
+
+    @staticmethod
+    def _stored(index: int) -> UploadedStoredFile:
+        return UploadedStoredFile(
+            asset=SimpleNamespace(id=f"asset-{index}"),
+            version=SimpleNamespace(id=f"version-{index}"),
+            blob=SimpleNamespace(id=f"blob-{index}"),
+        )
+
+    @staticmethod
+    def _response(index: int, action: str) -> FileAssetResponse:
+        now = datetime(2026, 8, 21, tzinfo=UTC).isoformat()
+        return FileAssetResponse(
+            id=f"asset-{index}",
+            tenant_id="tenant-1",
+            owner_type="user",
+            owner_id="user-1",
+            display_path=f"Imports/file-{index}.txt",
+            filename=f"file-{index}.txt",
+            size_bytes=index,
+            checksum_sha256=str(index) * 64,
+            version_id=f"version-{index}",
+            created_at=now,
+            updated_at=now,
+            source_revision=f"revision-{action}",
+        )
+
+    @patch("govoplan_files.backend.routes.connector_io.audit_from_principal")
+    @patch("govoplan_files.backend.routes.connector_io._asset_response")
+    @patch("govoplan_files.backend.routes.connector_io._audit_connector_sync")
+    @patch("govoplan_files.backend.routes.connector_io.sync_file_asset_from_source")
+    @patch("govoplan_files.backend.routes.connector_io._download_connector_payload")
+    @patch("govoplan_files.backend.routes.connector_io.discover_connector_folder")
+    @patch("govoplan_files.backend.routes.connector_io._connector_space_policy_decision")
+    @patch("govoplan_files.backend.routes.connector_io._visible_connector_profile")
+    @patch("govoplan_files.backend.routes.connector_io.get_connector_space_for_user")
+    @patch("govoplan_files.backend.routes.connector_io._is_admin", return_value=False)
+    def test_returns_reviewable_per_item_results_and_audits_the_summary(
+        self,
+        _is_admin: Mock,
+        get_space: Mock,
+        visible_profile: Mock,
+        policy_decision: Mock,
+        discover: Mock,
+        download: Mock,
+        sync_file: Mock,
+        audit_sync: Mock,
+        asset_response: Mock,
+        audit: Mock,
+    ) -> None:
+        get_space.return_value = self.space
+        visible_profile.return_value = self.profile
+        policy_decision.return_value = SimpleNamespace(allowed=True)
+        sources = [_item("file", f"linked/root/file-{index}.txt") for index in range(1, 6)]
+        discover.return_value = SimpleNamespace(
+            files=sources,
+            skipped=[SimpleNamespace(path="linked/root/ignored", reason="Unsupported item")],
+            truncated=False,
+        )
+        downloads = [
+            (
+                source.path,
+                ConnectorDownloadedFile(
+                    filename=source.name,
+                    data=source.name.encode(),
+                    revision=source.etag,
+                ),
+                {"source_provenance": {"connector_id": "profile-1"}},
+            )
+            for source in sources
+        ]
+        denied_decision = SimpleNamespace(
+            reason="Path denied",
+            to_dict=Mock(return_value={"allowed": False, "reason": "Path denied"}),
+        )
+        downloads[-1] = ConnectorPolicyDenied(denied_decision)  # type: ignore[assignment]
+        download.side_effect = downloads
+        sync_file.side_effect = [
+            (self._stored(1), "created", None),
+            (self._stored(2), "updated", "previous-2"),
+            (self._stored(3), "unchanged", "previous-3"),
+            FileStorageError("Target file already exists: Imports/file-4.txt"),
+        ]
+        asset_response.side_effect = [
+            self._response(1, "created"),
+            self._response(2, "updated"),
+            self._response(3, "unchanged"),
+        ]
+
+        response = sync_connector_space_folder(
+            "space-1",
+            FileConnectorFolderSyncRequest(target_folder="Imports"),
+            session=self.session,
+            principal=self.principal,  # type: ignore[arg-type]
+        )
+
+        self.assertEqual(
+            {
+                "discovered": 5,
+                "created": 1,
+                "updated": 1,
+                "unchanged": 1,
+                "skipped": 1,
+                "conflicts": 1,
+                "policy_denied": 1,
+                "failed": 0,
+            },
+            response.summary.model_dump(),
+        )
+        self.assertEqual(
+            [
+                "skipped",
+                "created",
+                "updated",
+                "unchanged",
+                "conflict",
+                "policy_denied",
+            ],
+            [item.action for item in response.items],
+        )
+        self.assertEqual("Imports/file-1.txt", response.items[1].target_path)
+        self.assertEqual(
+            {"allowed": False, "reason": "Path denied"},
+            response.items[-1].policy_decision,
+        )
+        self.session.commit.assert_called_once_with()
+        audit_sync.assert_has_calls(
+            [
+                call(
+                    self.session,
+                    self.principal,
+                    self._stored(index).asset,
+                    sync_action=action,
+                    previous_version_id=previous,
+                )
+                for index, action, previous in (
+                    (1, "created", None),
+                    (2, "updated", "previous-2"),
+                    (3, "unchanged", "previous-3"),
+                )
+            ]
+        )
+        summary_details = audit.call_args.kwargs["details"]
+        self.assertEqual(response.summary.model_dump(), summary_details["summary"])
+        self.assertEqual(6, len(summary_details["results"]))
+        self.assertNotIn("detail", summary_details["results"][-1])
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_connector_policy.py

@@ -0,0 +1,66 @@
+from __future__ import annotations
+
+import unittest
+
+from govoplan_files.backend.storage.connector_policy import (
+    ConnectorAccessRequest,
+    ConnectorPolicySource,
+    connector_policy_decision,
+)
+
+
+class ConnectorPolicyPatternTests(unittest.TestCase):
+    def test_resource_reference_allow_patterns_match(self) -> None:
+        decision = connector_policy_decision(
+            ConnectorAccessRequest(
+                connector_id="dev-smb",
+                credential_id="credential-primary",
+                provider="smb",
+            ),
+            (
+                ConnectorPolicySource(
+                    scope_type="tenant",
+                    scope_id="tenant-1",
+                    label="Tenant",
+                    policy={
+                        "allow": {
+                            "connectors": ["dev-*"],
+                            "credentials": ["credential-*"],
+                            "providers": ["s?b"],
+                        }
+                    },
+                ),
+            ),
+        )
+
+        self.assertTrue(decision.allowed)
+
+    def test_resource_reference_deny_patterns_take_precedence(self) -> None:
+        decision = connector_policy_decision(
+            ConnectorAccessRequest(
+                connector_id="dev-smb",
+                credential_id="credential-legacy",
+                provider="smb",
+            ),
+            (
+                ConnectorPolicySource(
+                    scope_type="tenant",
+                    scope_id="tenant-1",
+                    label="Tenant",
+                    policy={
+                        "allow": {"connectors": ["dev-*"]},
+                        "deny": {"credentials": ["*-legacy"]},
+                    },
+                ),
+            ),
+        )
+
+        self.assertFalse(decision.allowed)
+        self.assertEqual(
+            ("connector_policy_denylist",),
+            decision.requirements,
+        )
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_connector_profile_store.py

@@ -1,5 +1,6 @@
 from __future__ import annotations
 
+import json
 import unittest
 from unittest.mock import patch
 
@@ -8,6 +9,7 @@ from sqlalchemy.orm import sessionmaker
 
 from govoplan_access.backend.db import models as access_models  # noqa: F401 - resolve Files user foreign keys
 from govoplan_core.db.base import Base
+from govoplan_core.security.credential_envelopes import CredentialEnvelope
 from govoplan_core.security.secrets import decrypt_secret, encrypt_secret
 from govoplan_files.backend.db.models import FileConnectorProfile
 from govoplan_files.backend.storage.common import FileStorageError
@@ -20,7 +22,13 @@ from govoplan_files.backend.storage.connector_profile_store import (
 class ConnectorProfileStoreUpdateTests(unittest.TestCase):
     def setUp(self) -> None:
         self.engine = create_engine("sqlite:///:memory:")
-        Base.metadata.create_all(self.engine, tables=[FileConnectorProfile.__table__])
+        Base.metadata.create_all(
+            self.engine,
+            tables=[
+                FileConnectorProfile.__table__,
+                CredentialEnvelope.__table__,
+            ],
+        )
         self.session = sessionmaker(bind=self.engine)()
         self.row = FileConnectorProfile(
             id="profile-1",
@@ -171,6 +179,50 @@ class ConnectorProfileStoreUpdateTests(unittest.TestCase):
         self.assertEqual([self.row.id], [profile.id for profile in profiles])
         self.assertEqual("original-password", profiles[0].password_value)
 
+    def test_reusable_credential_resolves_and_revocation_keeps_profile_visible(self) -> None:
+        credential = CredentialEnvelope(
+            id="shared-credential",
+            tenant_id="tenant-1",
+            scope_type="tenant",
+            scope_id="tenant-1",
+            name="Shared WebDAV login",
+            credential_kind="username_password",
+            public_data={"username": "ada"},
+            secret_data_encrypted=encrypt_secret(json.dumps({"password": "secret"})),
+            secret_keys=["password"],
+            allowed_modules=["files"],
+            allowed_server_refs=[],
+            inherit_to_lower_scopes=True,
+            is_active=True,
+            revision="revision-1",
+        )
+        self.row.credential_profile_id = "credential-envelope:shared-credential"
+        self.row.username = None
+        self.row.password_encrypted = None
+        self.row.token_encrypted = None
+        self.session.add(credential)
+        self.session.commit()
+
+        profile = list_database_connector_profiles(
+            self.session,
+            tenant_id="tenant-1",
+        )[0]
+
+        self.assertEqual("ada", profile.username)
+        self.assertEqual("secret", profile.password_value)
+        self.assertTrue(profile.credentials_configured)
+
+        credential.is_active = False
+        self.session.commit()
+        profile = list_database_connector_profiles(
+            self.session,
+            tenant_id="tenant-1",
+        )[0]
+
+        self.assertIsNone(profile.password_value)
+        self.assertFalse(profile.credentials_configured)
+        self.assertTrue(profile.metadata["credential_unavailable"])
+
 
 if __name__ == "__main__":
     unittest.main()

tests/test_connector_providers.py

@@ -5,7 +5,7 @@ from datetime import UTC, datetime
 from unittest.mock import patch
 
 from govoplan_files.backend.storage.connector_browse import ConnectorBrowseError, _smb_location, browse_connector_profile
-from govoplan_files.backend.storage.connector_imports import ConnectorImportError, read_connector_file
+from govoplan_files.backend.storage.connector_imports import read_connector_file
 from govoplan_files.backend.storage.connector_profiles import ConnectorProfile, connector_profiles_from_payload
 from govoplan_files.backend.storage.connector_providers import connector_provider_descriptors
 
@@ -23,6 +23,10 @@ class FakeS3Client:
         self.list_objects_request: dict[str, object] | None = None
         self.head_request: dict[str, object] | None = None
         self.get_request: dict[str, object] | None = None
+        self.closed = False
+
+    def close(self) -> None:
+        self.closed = True
 
     def list_objects_v2(self, **kwargs: object) -> dict[str, object]:
         self.list_objects_request = dict(kwargs)
@@ -75,42 +79,39 @@ def s3_profile(**overrides: object) -> ConnectorProfile:
 
 
 class ConnectorProviderTests(unittest.TestCase):
-    def test_smb_sdk_transport_fails_closed_before_client_creation_in_all_modes(self) -> None:
+    def test_smb_browse_uses_the_files_owned_pinned_connection_cache(self) -> None:
         profile = ConnectorProfile(
             id="smb",
             label="SMB",
             provider="smb",
             endpoint_url="smb://files.example.test/share",
         )
-        for allow_private, address in ((False, "93.184.216.34"), (True, "10.0.0.5")):
+        with patch.dict(
-            with self.subTest(allow_private=allow_private), patch.dict(
+            "os.environ",
-                "os.environ",
+            {"APP_ENV": "production", "GOVOPLAN_CONNECTOR_ALLOW_PRIVATE_NETWORKS": "false"},
-                {
+        ), patch(
-                    "APP_ENV": "production",
+            "govoplan_core.security.outbound_http.socket.getaddrinfo",
-                    "GOVOPLAN_CONNECTOR_ALLOW_PRIVATE_NETWORKS": str(allow_private).lower(),
+            return_value=[(2, 1, 6, "", ("93.184.216.34", 445))],
-                },
+        ), patch("govoplan_files.backend.storage.connector_browse._smbclient_module") as sdk:
-            ), patch(
+            sdk.return_value.scandir.return_value.__enter__.return_value = iter(())
-                "govoplan_core.security.outbound_http.socket.getaddrinfo",
+            self.assertEqual([], browse_connector_profile(profile, path=""))
-                return_value=[(2, 1, 6, "", (address, 445))],
-            ), patch("govoplan_files.backend.storage.connector_browse._smbclient_module") as sdk, self.assertRaisesRegex(
-                ConnectorBrowseError,
-                "redirects/referrals.*DNS/IP pinning",
-            ):
-                browse_connector_profile(profile, path="")
-            sdk.assert_not_called()
 
-    def test_smb_explicit_ip_still_fails_closed_because_the_sdk_may_follow_referrals(self) -> None:
+        kwargs = sdk.return_value.scandir.call_args.kwargs
+        self.assertIsInstance(kwargs["connection_cache"], dict)
+        self.assertTrue(kwargs["require_signing"])
+
+    def test_smb_endpoint_preflight_applies_private_network_policy(self) -> None:
         profile = ConnectorProfile(id="smb", label="SMB", provider="smb", endpoint_url="smb://10.0.0.5/share")
         with patch.dict(
             "os.environ",
-            {"APP_ENV": "production", "GOVOPLAN_CONNECTOR_ALLOW_PRIVATE_NETWORKS": "true"},
+            {"APP_ENV": "production", "GOVOPLAN_CONNECTOR_ALLOW_PRIVATE_NETWORKS": "false"},
         ), patch(
             "govoplan_core.security.outbound_http.socket.getaddrinfo",
             return_value=[(2, 1, 6, "", ("10.0.0.5", 445))],
-        ), self.assertRaisesRegex(ConnectorBrowseError, "redirects/referrals.*DNS/IP pinning"):
+        ), self.assertRaisesRegex(ConnectorBrowseError, "non-public network"):
             _smb_location(profile)
 
-    def test_smb_import_surfaces_fail_closed_policy_as_an_import_error(self) -> None:
+    def test_smb_import_uses_the_same_pinned_connection_cache(self) -> None:
         profile = ConnectorProfile(id="smb", label="SMB", provider="smb", endpoint_url="smb://10.0.0.5/share")
         with patch.dict(
             "os.environ",
@@ -118,12 +119,13 @@ class ConnectorProviderTests(unittest.TestCase):
         ), patch(
             "govoplan_core.security.outbound_http.socket.getaddrinfo",
             return_value=[(2, 1, 6, "", ("10.0.0.5", 445))],
-        ), patch("govoplan_files.backend.storage.connector_imports._smbclient_module") as sdk, self.assertRaisesRegex(
+        ), patch("govoplan_files.backend.storage.connector_imports._smbclient_module") as sdk:
-            ConnectorImportError,
+            sdk.return_value.stat.return_value.st_size = 4
-            "redirects/referrals.*DNS/IP pinning",
+            sdk.return_value.open_file.return_value.__enter__.return_value.read.return_value = b"test"
-        ):
+            downloaded = read_connector_file(profile, library_id="", path="notice.txt", max_bytes=1024)
-            read_connector_file(profile, library_id="", path="notice.txt", max_bytes=1024)
+
-        sdk.assert_not_called()
+        self.assertEqual(b"test", downloaded.data)
+        self.assertIsInstance(sdk.return_value.stat.call_args.kwargs["connection_cache"], dict)
 
     def test_provider_descriptors_include_s3_and_reserved_microsoft_providers(self) -> None:
         descriptors = {descriptor.provider: descriptor for descriptor in connector_provider_descriptors()}
@@ -172,6 +174,7 @@ class ConnectorProviderTests(unittest.TestCase):
         self.assertEqual("govoplan:root/report.xlsx", items[1].external_id)
         self.assertEqual("root/report.xlsx", items[1].metadata["key"])
         self.assertEqual("next-page", items[1].metadata["next_continuation_token"])
+        self.assertTrue(client.closed)
 
     def test_s3_browse_lists_buckets_when_profile_has_no_bucket(self) -> None:
         client = FakeS3Client()
@@ -181,27 +184,33 @@ class ConnectorProviderTests(unittest.TestCase):
 
         self.assertEqual(["archive"], [item.path for item in items])
 
-    def test_s3_sdk_transport_fails_closed_before_client_creation_in_private_mode(self) -> None:
+    def test_s3_client_is_constructed_through_the_pinned_transport(self) -> None:
         with patch.dict(
             "os.environ",
             {"APP_ENV": "production", "GOVOPLAN_CONNECTOR_ALLOW_PRIVATE_NETWORKS": "true"},
         ), patch(
             "govoplan_core.security.outbound_http.socket.getaddrinfo",
             return_value=[(2, 1, 6, "", ("127.0.0.1", 9000))],
-        ), patch("govoplan_files.backend.storage.connector_browse.import_module") as importer, self.assertRaisesRegex(
+        ), patch(
-            ConnectorBrowseError,
+            "govoplan_files.backend.storage.connector_browse.create_pinned_s3_client",
-            "until that transport supports.*DNS/IP pinning",
+            return_value=FakeS3Client(),
-        ):
+        ) as factory:
             browse_connector_profile(s3_profile(), path="")
-        importer.assert_not_called()
 
-    def test_s3_sdk_endpoint_discovery_fails_closed(self) -> None:
+        kwargs = factory.call_args.kwargs
-        with patch("govoplan_files.backend.storage.connector_browse.import_module") as importer, self.assertRaisesRegex(
+        self.assertEqual("http://127.0.0.1:9000", kwargs["endpoint_url"])
-            ConnectorBrowseError,
+        self.assertEqual("access-key", kwargs["aws_access_key_id"])
-            "endpoint discovery.*cannot guarantee.*DNS/IP pinning",
+        self.assertEqual("secret-key", kwargs["aws_secret_access_key"])
-        ):
+        self.assertEqual({}, kwargs["config"].proxies)
+
+    def test_s3_endpoint_discovery_uses_the_same_pinned_transport(self) -> None:
+        with patch(
+            "govoplan_files.backend.storage.connector_browse.create_pinned_s3_client",
+            return_value=FakeS3Client(),
+        ) as factory:
             browse_connector_profile(s3_profile(endpoint_url=None), path="")
-        importer.assert_not_called()
+
+        self.assertNotIn("endpoint_url", factory.call_args.kwargs)
 
     def test_s3_import_downloads_object_and_preserves_remote_identity(self) -> None:
         client = FakeS3Client()
@@ -217,6 +226,7 @@ class ConnectorProviderTests(unittest.TestCase):
         self.assertEqual("govoplan:root/report.txt", downloaded.external_id)
         self.assertEqual("s3://govoplan/root/report.txt", downloaded.external_url)
         self.assertEqual("checksum", downloaded.metadata["checksum_sha256"])
+        self.assertTrue(client.closed)
 
 
 if __name__ == "__main__":

tests/test_connector_space_deletion.py

@@ -0,0 +1,101 @@
+from __future__ import annotations
+
+import unittest
+from types import SimpleNamespace
+from unittest.mock import Mock, patch
+
+from govoplan_files.backend.storage.common import FileStorageError
+from govoplan_files.backend.storage.connector_profiles import ConnectorProfile
+from govoplan_files.backend.storage.connector_spaces import (
+    soft_delete_connector_space,
+    validate_connector_space_write_mode,
+)
+
+
+class ConnectorSpaceDeletionTests(unittest.TestCase):
+    def setUp(self) -> None:
+        self.session = Mock()
+        self.space = SimpleNamespace(
+            tenant_id="tenant-1",
+            owner_type="user",
+            owner_user_id="owner-1",
+            owner_group_id=None,
+            deleted_at=None,
+            is_active=True,
+        )
+
+    @patch("govoplan_files.backend.storage.connector_spaces.utcnow")
+    @patch("govoplan_files.backend.storage.connector_spaces.ensure_owner_access")
+    def test_owner_can_soft_delete_local_space_definition(
+        self, ensure_owner_access: Mock, utcnow: Mock
+    ) -> None:
+        deleted_at = object()
+        utcnow.return_value = deleted_at
+
+        result = soft_delete_connector_space(
+            self.session, self.space, user_id="owner-1"
+        )
+
+        self.assertIs(result, self.space)
+        self.assertIs(deleted_at, self.space.deleted_at)
+        self.assertFalse(self.space.is_active)
+        ensure_owner_access.assert_called_once_with(
+            self.session,
+            tenant_id="tenant-1",
+            owner_type="user",
+            owner_id="owner-1",
+            user_id="owner-1",
+            is_admin=False,
+        )
+        self.session.add.assert_called_once_with(self.space)
+        self.session.flush.assert_called_once_with()
+
+    @patch(
+        "govoplan_files.backend.storage.connector_spaces.ensure_owner_access",
+        side_effect=FileStorageError("File space access denied"),
+    )
+    def test_inaccessible_space_is_blocked_without_mutation(
+        self, _ensure_owner_access: Mock
+    ) -> None:
+        with self.assertRaisesRegex(FileStorageError, "access denied"):
+            soft_delete_connector_space(
+                self.session, self.space, user_id="other-user"
+            )
+
+        self.assertIsNone(self.space.deleted_at)
+        self.assertTrue(self.space.is_active)
+        self.session.add.assert_not_called()
+        self.session.flush.assert_not_called()
+
+    def test_read_only_mode_does_not_require_remote_write_capability(self) -> None:
+        profile = ConnectorProfile(
+            id="webdav-read",
+            label="Read only",
+            provider="webdav",
+            capabilities=("browse", "import"),
+        )
+
+        validate_connector_space_write_mode(profile, read_only=True)
+
+    def test_two_way_mode_requires_explicit_supported_write_capability(self) -> None:
+        missing_capability = ConnectorProfile(
+            id="s3-read",
+            label="S3 read only",
+            provider="s3",
+            capabilities=("browse",),
+        )
+        unsupported_provider = ConnectorProfile(
+            id="webdav-write",
+            label="WebDAV write",
+            provider="webdav",
+            capabilities=("write",),
+        )
+
+        for profile in (missing_capability, unsupported_provider):
+            with self.subTest(profile=profile.id):
+                with self.assertRaisesRegex(FileStorageError, "requires an S3"):
+                    validate_connector_space_write_mode(profile, read_only=False)
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_connector_visibility.py

@@ -10,6 +10,9 @@ from govoplan_files.backend.storage.connector_visibility import (
     connector_profile_usable_for_import,
     visible_connector_profiles_for_actor,
 )
+from govoplan_files.backend.storage.connector_providers import (
+    connector_provider_descriptors,
+)
 
 
 def _profile(
@@ -168,12 +171,16 @@ class ConnectorVisibilityTests(unittest.TestCase):
         self,
     ) -> None:
         self.assertTrue(connector_profile_usable_for_import(_profile("webdav")))
-        self.assertFalse(
+        descriptors = {
-            connector_profile_usable_for_import(_profile("s3", provider="s3"))
+            item.provider: item for item in connector_provider_descriptors()
-        )
+        }
-        self.assertFalse(
+        for provider in ("s3", "smb"):
-            connector_profile_usable_for_import(_profile("smb", provider="smb"))
+            self.assertEqual(
-        )
+                descriptors[provider].installed,
+                connector_profile_usable_for_import(
+                    _profile(provider, provider=provider)
+                ),
+            )
         self.assertFalse(
             connector_profile_usable_for_import(
                 _profile("sharepoint", provider="sharepoint")

tests/test_content_protection.py

@@ -0,0 +1,149 @@
+from __future__ import annotations
+
+from datetime import UTC, datetime
+import hashlib
+import unittest
+
+from sqlalchemy import create_engine
+from sqlalchemy.orm import Session
+
+from govoplan_core.core.encryption import (
+    CAPABILITY_ENCRYPTION_CONTENT_CIPHER,
+    ContentProtectionEnvelope,
+    ProtectedContent,
+)
+from govoplan_files.backend.db.models import FileBlob
+from govoplan_files.backend.runtime import configure_runtime
+from govoplan_files.backend.storage.common import FileStorageError
+from govoplan_files.backend.storage.content_protection import protect_blob_content
+from govoplan_files.backend.storage.integrity import read_verified_blob_bytes
+
+
+class Registry:
+    def __init__(self, capability=None) -> None:
+        self.capability_value = capability
+
+    def has_capability(self, name: str) -> bool:
+        return (
+            name == CAPABILITY_ENCRYPTION_CONTENT_CIPHER
+            and self.capability_value is not None
+        )
+
+    def capability(self, _name: str):
+        return self.capability_value
+
+
+class Cipher:
+    def protect_content(self, _session, *, request):
+        ciphertext = b"protected:" + request.plaintext
+        envelope = ContentProtectionEnvelope(
+            envelope_id="envelope-file-1",
+            tenant_id=request.tenant_id,
+            owner_module=request.owner_module,
+            resource_type=request.resource_type,
+            resource_id=request.resource_id,
+            profile_kind="server_envelope",
+            profile_id=request.profile_id,
+            provider_id="test",
+            vault_id=request.vault_id,
+            key_version=1,
+            algorithm_suite="AES-256-GCM",
+            ciphertext_ref=request.ciphertext_ref,
+            ciphertext_digest=f"sha256:{hashlib.sha256(ciphertext).hexdigest()}",
+            authenticated_context_digest="sha256:context",
+            state="active",
+            created_at=datetime.now(tz=UTC),
+            wrapped_key_refs=("wrapped:1",),
+        )
+        return ProtectedContent(envelope=envelope, ciphertext=ciphertext)
+
+    def unprotect_content(self, _session, *, request):
+        if not request.ciphertext.startswith(b"protected:"):
+            raise ValueError("invalid fixture ciphertext")
+        return request.ciphertext.removeprefix(b"protected:")
+
+    def execute_rewrap(self, *_args, **_kwargs):
+        raise NotImplementedError
+
+    def prepare_reencryption(self, *_args, **_kwargs):
+        raise NotImplementedError
+
+
+class Backend:
+    def __init__(self, data: bytes) -> None:
+        self.data = data
+
+    def get_bytes(self, _key: str) -> bytes:
+        return self.data
+
+
+class FileContentProtectionTests(unittest.TestCase):
+    def setUp(self) -> None:
+        self.engine = create_engine("sqlite+pysqlite:///:memory:")
+        FileBlob.__table__.create(self.engine)
+        self.session = Session(self.engine)
+
+    def tearDown(self) -> None:
+        self.session.close()
+        self.engine.dispose()
+
+    def test_encrypted_blob_is_opened_after_stored_object_integrity_check(self) -> None:
+        configure_runtime(registry=Registry(Cipher()), settings=object())
+        plaintext = b"monthly records"
+        protected = protect_blob_content(
+            self.session,
+            tenant_id="tenant-1",
+            blob_id="blob-1",
+            vault_id="vault-1",
+            ciphertext_ref="tenants/tenant-1/files/blob-1",
+            plaintext=plaintext,
+            actor_id="account-1",
+            content_type="text/plain",
+        )
+        blob = FileBlob(
+            id="blob-1",
+            tenant_id="tenant-1",
+            storage_backend="test",
+            storage_key="tenants/tenant-1/files/blob-1",
+            checksum_sha256=hashlib.sha256(plaintext).hexdigest(),
+            size_bytes=len(plaintext),
+            protection_discriminator="vault:vault-1",
+            encryption_envelope_id=protected.envelope.envelope_id,
+            storage_checksum_sha256=hashlib.sha256(protected.ciphertext).hexdigest(),
+            storage_size_bytes=len(protected.ciphertext),
+            content_type="text/plain",
+            ref_count=1,
+            integrity_status="verified",
+        )
+        self.session.add(blob)
+        self.session.flush()
+        self.assertEqual(
+            plaintext,
+            read_verified_blob_bytes(blob, backend=Backend(protected.ciphertext)),
+        )
+
+    def test_encrypted_blob_fails_closed_without_encryption_capability(self) -> None:
+        configure_runtime(registry=Registry(), settings=object())
+        ciphertext = b"protected:monthly records"
+        blob = FileBlob(
+            id="blob-1",
+            tenant_id="tenant-1",
+            storage_backend="test",
+            storage_key="tenants/tenant-1/files/blob-1",
+            checksum_sha256=hashlib.sha256(b"monthly records").hexdigest(),
+            size_bytes=len(b"monthly records"),
+            protection_discriminator="vault:vault-1",
+            encryption_envelope_id="envelope-file-1",
+            storage_checksum_sha256=hashlib.sha256(ciphertext).hexdigest(),
+            storage_size_bytes=len(ciphertext),
+            ref_count=1,
+            integrity_status="verified",
+        )
+        self.session.add(blob)
+        self.session.flush()
+        with self.assertRaisesRegex(FileStorageError, "Encryption is unavailable"):
+            read_verified_blob_bytes(blob, backend=Backend(ciphertext))
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_documentation.py

@@ -49,6 +49,10 @@ class FilesRuntimeDocumentationTests(unittest.TestCase):
         settings = SimpleNamespace(
             file_upload_max_bytes=7 * 1024 * 1024,
             file_upload_zip_max_bytes=19 * 1024 * 1024,
+            file_archive_max_expanded_bytes=31 * 1024 * 1024,
+            file_archive_max_entries=321,
+            file_archive_max_expansion_ratio=42,
+            file_archive_preview_ttl_seconds=15 * 60,
         )
         with patch(
             "govoplan_files.backend.documentation.visible_connector_profiles_for_actor",
@@ -73,8 +77,12 @@ class FilesRuntimeDocumentationTests(unittest.TestCase):
 
         archive = topics["files.workflow.upload-and-unpack-zip"]
         self.assertIn("19 MiB (19,922,944 bytes)", archive.body)
+        self.assertIn("31 MiB (32,505,856 bytes)", archive.body)
         self.assertIn("7 MiB (7,340,032 bytes)", archive.body)
-        self.assertIn("1,000", archive.body)
+        self.assertIn("321", archive.body)
+        self.assertIn("42:1", archive.body)
+        self.assertIn("15 minutes", archive.body)
+        self.assertIn("passwords remain request-only", archive.body)
         self.assertIn("Actual extracted bytes are counted", archive.body)
 
     def test_connector_task_requires_authority_and_a_visible_usable_profile(
@@ -134,6 +142,7 @@ class FilesRuntimeDocumentationTests(unittest.TestCase):
             base_path="/classified",
             credential_mode="basic",
             password_value="credential-secret",
+            secret_ref="runtime/connector-secret",
         )
         with patch(
             "govoplan_files.backend.documentation.visible_connector_profiles_for_actor",

tests/test_dsar_provider.py

@@ -0,0 +1,468 @@
+from __future__ import annotations
+
+import unittest
+from datetime import datetime, timedelta, timezone
+
+from sqlalchemy import create_engine
+from sqlalchemy.orm import sessionmaker
+
+from govoplan_access.backend.db.models import Account, Group, User
+from govoplan_core.core.dsar import DsarProvider, DsarSubjectRef
+from govoplan_core.core.change_sequence import ChangeSequenceEntry
+from govoplan_core.db.base import Base
+from govoplan_core.privacy.dsar_workflow import (
+    DataSubjectRequest,
+    create_data_subject_request,
+    execute_data_subject_erasure,
+    plan_data_subject_erasure,
+    search_data_subject_request,
+)
+from govoplan_files.backend.db.models import (
+    FileAsset,
+    FileBlob,
+    FileConnectorCredential,
+    FileConnectorPolicy,
+    FileConnectorProfile,
+    FileConnectorSpace,
+    FileFolder,
+    FileFormEvidenceGrant,
+    FileIntegrityFinding,
+    FileIntegrityScan,
+    FileShare,
+    FileVersion,
+)
+from govoplan_files.backend.dsar_provider import (
+    FILES_DSAR_CAPABILITY,
+    FilesDsarProvider,
+)
+from govoplan_files.backend.manifest import manifest
+
+
+class _Registry:
+    def __init__(self, provider: FilesDsarProvider, *, files_active: bool = True) -> None:
+        self.provider = provider
+        self.files_active = files_active
+
+    def capability_names(self):
+        return (FILES_DSAR_CAPABILITY,)
+
+    def capability_owner(self, name):
+        self._assert_capability(name)
+        return "files"
+
+    def tenant_entitlement_resolver(self):
+        files_active = self.files_active
+
+        class _Resolver:
+            @staticmethod
+            def resolve(session, tenant_id):
+                del session, tenant_id
+                return type(
+                    "State",
+                    (),
+                    {"effective_modules": ("files",) if files_active else ()},
+                )()
+
+        return _Resolver()
+
+    def require_tenant_capability(self, name, session, **kwargs):
+        del session, kwargs
+        self._assert_capability(name)
+        return self.provider
+
+    def manifests(self):
+        return (type("Manifest", (), {"id": "files"})(),)
+
+    @staticmethod
+    def _assert_capability(name: str) -> None:
+        if name != FILES_DSAR_CAPABILITY:
+            raise KeyError(name)
+
+
+class FilesDsarProviderTests(unittest.TestCase):
+    def setUp(self) -> None:
+        self.engine = create_engine("sqlite:///:memory:", future=True)
+        Base.metadata.create_all(
+            bind=self.engine,
+            tables=[
+                Account.__table__,
+                User.__table__,
+                Group.__table__,
+                ChangeSequenceEntry.__table__,
+                DataSubjectRequest.__table__,
+                FileBlob.__table__,
+                FileAsset.__table__,
+                FileVersion.__table__,
+                FileFolder.__table__,
+                FileFormEvidenceGrant.__table__,
+                FileShare.__table__,
+                FileConnectorCredential.__table__,
+                FileConnectorPolicy.__table__,
+                FileConnectorProfile.__table__,
+                FileConnectorSpace.__table__,
+                FileIntegrityScan.__table__,
+                FileIntegrityFinding.__table__,
+            ],
+        )
+        self.session = sessionmaker(bind=self.engine, future=True)()
+        self.account = Account(
+            id="account-1",
+            email="subject@example.test",
+            normalized_email="subject@example.test",
+            display_name="Subject",
+            password_hash="not-exported",
+        )
+        self.user = User(
+            id="membership-1",
+            tenant_id="tenant-1",
+            account_id=self.account.id,
+            email="subject@example.test",
+            display_name="Subject",
+        )
+        self.blob = FileBlob(
+            id="blob-1",
+            tenant_id="tenant-1",
+            storage_backend="local",
+            storage_key="private/storage/key-do-not-export",
+            checksum_sha256="a" * 64,
+            size_bytes=12,
+            ref_count=1,
+        )
+        self.asset = FileAsset(
+            id="asset-1",
+            tenant_id="tenant-1",
+            owner_type="user",
+            owner_user_id=self.user.id,
+            created_by_user_id=self.user.id,
+            current_version_id="version-1",
+            display_path="subjects/private.txt",
+            filename="private.txt",
+            description="Subject-provided document",
+            retained_until=datetime.now(timezone.utc) + timedelta(days=30),
+            lifecycle_reason="Pending proceeding",
+            metadata_={"password": "metadata-secret-do-not-export"},
+        )
+        self.version = FileVersion(
+            id="version-1",
+            tenant_id="tenant-1",
+            file_asset_id=self.asset.id,
+            blob_id=self.blob.id,
+            version_number=1,
+            filename_at_upload=self.asset.filename,
+            display_path_at_upload=self.asset.display_path,
+            content_type="text/plain",
+            size_bytes=12,
+            checksum_sha256="a" * 64,
+            created_by_user_id=self.user.id,
+        )
+        self.folder = FileFolder(
+            id="folder-1",
+            tenant_id="tenant-1",
+            owner_type="user",
+            owner_user_id=self.user.id,
+            path="subjects",
+            created_by_user_id=self.user.id,
+        )
+        self.share = FileShare(
+            id="share-1",
+            tenant_id="tenant-1",
+            file_asset_id=self.asset.id,
+            target_type="user",
+            target_id=self.user.id,
+            permission="read",
+            created_by_user_id=self.user.id,
+        )
+        self.form_evidence = FileFormEvidenceGrant(
+            id="evidence-1",
+            tenant_id="tenant-1",
+            form_instance_id="form-instance-1",
+            form_definition_id="application",
+            form_definition_revision="7",
+            token_sha256="b" * 64,
+            idempotency_key="evidence-key-1",
+            request_sha256="c" * 64,
+            custodian_user_id=self.user.id,
+            evidence_kind="attachment",
+            purpose="Submitted application evidence",
+            status="uploaded",
+            expires_at=datetime.now(timezone.utc) + timedelta(days=1),
+            max_size_bytes=1024,
+            allowed_content_types=["text/plain"],
+            file_asset_id=self.asset.id,
+            file_version_id=self.version.id,
+            metadata_={},
+        )
+        self.credential = FileConnectorCredential(
+            id="credential-1",
+            tenant_id="tenant-1",
+            scope_type="tenant",
+            scope_id="tenant-1",
+            label="Subject-created credential",
+            provider="s3",
+            credential_mode="database",
+            username="subject-user",
+            password_encrypted="encrypted-password-do-not-export",
+            token_encrypted="encrypted-token-do-not-export",
+            password_env="PASSWORD_ENV_DO_NOT_EXPORT",
+            secret_ref="vault://do-not-export",
+            created_by_user_id=self.user.id,
+            updated_by_user_id=self.user.id,
+        )
+        self.integrity_scan = FileIntegrityScan(
+            id="scan-1",
+            tenant_id="tenant-1",
+            storage_backend="local",
+            storage_prefix="private-prefix-do-not-export",
+            created_by_user_id=self.user.id,
+        )
+        tenant_two_asset = FileAsset(
+            id="asset-tenant-2",
+            tenant_id="tenant-2",
+            owner_type="user",
+            owner_user_id=self.user.id,
+            display_path="other-tenant.txt",
+            filename="other-tenant.txt",
+        )
+        self.session.add_all(
+            [
+                self.account,
+                self.user,
+                self.blob,
+                self.asset,
+                self.version,
+                self.folder,
+                self.share,
+                self.form_evidence,
+                self.credential,
+                self.integrity_scan,
+                tenant_two_asset,
+            ]
+        )
+        self.session.commit()
+        self.provider = FilesDsarProvider()
+        self.subject = DsarSubjectRef(membership_id=self.user.id)
+
+    def tearDown(self) -> None:
+        self.session.close()
+        self.engine.dispose()
+
+    def test_manifest_publishes_protocol_conforming_provider(self) -> None:
+        provided_names = {item.name for item in manifest.provides_interfaces}
+        self.assertIn(FILES_DSAR_CAPABILITY, provided_names)
+        provider = manifest.capability_factories[FILES_DSAR_CAPABILITY](None)
+        self.assertIsInstance(provider, DsarProvider)
+
+    def test_search_is_tenant_scoped_and_excludes_bytes_and_secrets(self) -> None:
+        records = self.provider.search_subject(
+            self.session,
+            tenant_id="tenant-1",
+            subject=self.subject,
+        )
+        resource_types = {record.resource_type for record in records}
+        self.assertTrue(
+            {
+                "file_asset",
+                "file_version",
+                "file_folder",
+                "file_share",
+                "file_form_evidence",
+                "connector_credential",
+                "file_integrity_scan",
+            }.issubset(resource_types)
+        )
+        serialized = repr([record.to_dict() for record in records])
+        self.assertNotIn("asset-tenant-2", serialized)
+        self.assertNotIn("private/storage/key-do-not-export", serialized)
+        self.assertNotIn("metadata-secret-do-not-export", serialized)
+        self.assertNotIn("encrypted-password-do-not-export", serialized)
+        self.assertNotIn("encrypted-token-do-not-export", serialized)
+        self.assertNotIn("subject-user", serialized)
+        self.assertNotIn("PASSWORD_ENV_DO_NOT_EXPORT", serialized)
+        self.assertNotIn("vault://do-not-export", serialized)
+        self.assertNotIn("private-prefix-do-not-export", serialized)
+
+    def test_conflicting_direct_subject_references_fail_closed(self) -> None:
+        records = self.provider.search_subject(
+            self.session,
+            tenant_id="tenant-1",
+            subject=DsarSubjectRef(
+                membership_id=self.user.id,
+                external_references={"files.user": "another-membership"},
+            ),
+        )
+
+        self.assertEqual((), records)
+
+    def test_plan_classifies_retention_and_manual_review(self) -> None:
+        records = self.provider.search_subject(
+            self.session,
+            tenant_id="tenant-1",
+            subject=self.subject,
+        )
+        actions = self.provider.plan_erasure(
+            self.session,
+            tenant_id="tenant-1",
+            subject=self.subject,
+            records=records,
+        )
+
+        kinds = {action.kind for action in actions}
+        self.assertTrue({"retain", "manual_review", "revoke", "detach"}.issubset(kinds))
+        asset_retention = next(
+            action
+            for action in actions
+            if action.action_id == "files:retain:file_asset:asset-1"
+        )
+        self.assertIn("retained until", asset_retention.rationale)
+        self.assertTrue(
+            any(action.action_id == "files:revoke:file_share:share-1" for action in actions)
+        )
+        self.assertFalse(
+            any(action.kind == "delete" and action.executable for action in actions)
+        )
+
+    def test_execution_is_revalidated_and_idempotent(self) -> None:
+        records = self.provider.search_subject(
+            self.session,
+            tenant_id="tenant-1",
+            subject=self.subject,
+        )
+        actions = self.provider.plan_erasure(
+            self.session,
+            tenant_id="tenant-1",
+            subject=self.subject,
+            records=records,
+        )
+        executable = tuple(action for action in actions if action.executable)
+
+        first = self.provider.execute_erasure(
+            self.session,
+            tenant_id="tenant-1",
+            subject=self.subject,
+            actions=executable,
+            request_id="dsar-1",
+        )
+        self.assertEqual({"executed"}, {result.status for result in first})
+        self.assertIsNotNone(self.share.revoked_at)
+        self.assertIsNone(self.asset.created_by_user_id)
+        self.assertIsNone(self.version.created_by_user_id)
+        self.assertIsNone(self.folder.created_by_user_id)
+        self.assertIsNone(self.credential.created_by_user_id)
+        self.assertIsNone(self.credential.updated_by_user_id)
+
+        repeated = self.provider.execute_erasure(
+            self.session,
+            tenant_id="tenant-1",
+            subject=self.subject,
+            actions=executable,
+            request_id="dsar-1",
+        )
+        self.assertEqual({"unchanged"}, {result.status for result in repeated})
+
+    def test_execution_blocks_when_reference_changed_after_planning(self) -> None:
+        records = self.provider.search_subject(
+            self.session,
+            tenant_id="tenant-1",
+            subject=self.subject,
+        )
+        actions = self.provider.plan_erasure(
+            self.session,
+            tenant_id="tenant-1",
+            subject=self.subject,
+            records=records,
+        )
+        detach = next(
+            action
+            for action in actions
+            if action.action_id
+            == "files:detach:file_version:created_by_user_id:version-1"
+        )
+        self.version.created_by_user_id = "replacement-user"
+        self.session.flush()
+
+        result = self.provider.execute_erasure(
+            self.session,
+            tenant_id="tenant-1",
+            subject=self.subject,
+            actions=(detach,),
+            request_id="dsar-2",
+        )
+
+        self.assertEqual("blocked", result[0].status)
+        self.assertEqual("replacement-user", self.version.created_by_user_id)
+
+    def test_core_workflow_discovers_active_provider_and_skips_it_when_disabled(
+        self,
+    ) -> None:
+        request = create_data_subject_request(
+            self.session,
+            tenant_id="tenant-1",
+            reference="DSAR-FILES-1",
+            request_kind="access_and_erasure",
+            subject=self.subject,
+            purpose="Respond to an authorized privacy request.",
+            legal_basis="Article 15 and 17 GDPR",
+            due_at=None,
+            requested_by_account_id="privacy-officer",
+        )
+        self.session.commit()
+        registry = _Registry(self.provider)
+
+        search_data_subject_request(
+            self.session,
+            registry=registry,
+            row=request,
+            expected_revision=1,
+        )
+        self.assertEqual(
+            "searched", request.status, request.search_result["provider_runs"]
+        )
+        self.assertEqual(["files"], request.coverage["covered_modules"])
+        self.assertEqual([], request.coverage["modules_without_provider"])
+        plan_data_subject_erasure(
+            self.session,
+            registry=registry,
+            row=request,
+            expected_revision=2,
+        )
+        executable_ids = [
+            action["action_id"]
+            for action in request.erasure_plan["actions"]
+            if action["executable"]
+        ]
+        execute_data_subject_erasure(
+            self.session,
+            registry=registry,
+            row=request,
+            expected_revision=3,
+            action_ids=executable_ids,
+        )
+        self.assertEqual("completed", request.status)
+
+        disabled = create_data_subject_request(
+            self.session,
+            tenant_id="tenant-1",
+            reference="DSAR-FILES-DISABLED",
+            request_kind="access",
+            subject=self.subject,
+            purpose="Verify disabled-module coverage.",
+            legal_basis="Article 15 GDPR",
+            due_at=None,
+            requested_by_account_id="privacy-officer",
+        )
+        search_data_subject_request(
+            self.session,
+            registry=_Registry(self.provider, files_active=False),
+            row=disabled,
+            expected_revision=1,
+        )
+
+        self.assertEqual(0, disabled.search_result["record_count"])
+        self.assertEqual(
+            [FILES_DSAR_CAPABILITY],
+            disabled.coverage["inactive_provider_capabilities"],
+        )
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_file_lifecycle.py

@@ -0,0 +1,164 @@
+from __future__ import annotations
+
+from datetime import timedelta
+import unittest
+
+from sqlalchemy import create_engine
+from sqlalchemy.orm import sessionmaker
+
+from govoplan_access.backend.db.models import Account, Group, User
+from govoplan_core.db.base import Base
+from govoplan_core.core.change_sequence import ChangeSequenceEntry
+from govoplan_files.backend.db.models import (
+    FileAsset,
+    FileBlob,
+    FileFolder,
+    FileFormEvidenceGrant,
+    FileShare,
+    FileVersion,
+)
+from govoplan_files.backend.storage.common import FileStorageError, utcnow
+from govoplan_files.backend.storage.lifecycle import (
+    preview_asset_purge,
+    restore_asset,
+    set_asset_lifecycle,
+)
+
+
+class FileLifecycleTests(unittest.TestCase):
+    def setUp(self) -> None:
+        self.engine = create_engine("sqlite:///:memory:", future=True)
+        Base.metadata.create_all(
+            bind=self.engine,
+            tables=[
+                Account.__table__,
+                User.__table__,
+                Group.__table__,
+                ChangeSequenceEntry.__table__,
+                FileBlob.__table__,
+                FileAsset.__table__,
+                FileVersion.__table__,
+                FileFolder.__table__,
+                FileFormEvidenceGrant.__table__,
+                FileShare.__table__,
+            ],
+        )
+        self.session = sessionmaker(bind=self.engine, future=True)()
+        self.blob = FileBlob(
+            id="blob-1",
+            tenant_id="tenant-1",
+            storage_backend="local",
+            storage_key="tenants/tenant-1/files/blob-1",
+            checksum_sha256="a" * 64,
+            size_bytes=1,
+            ref_count=1,
+        )
+        self.asset = FileAsset(
+            id="file-1",
+            tenant_id="tenant-1",
+            owner_type="user",
+            owner_user_id="user-1",
+            current_version_id="version-1",
+            display_path="records/report.txt",
+            filename="report.txt",
+            deleted_at=utcnow(),
+        )
+        self.version = FileVersion(
+            id="version-1",
+            tenant_id="tenant-1",
+            file_asset_id="file-1",
+            blob_id="blob-1",
+            version_number=1,
+            filename_at_upload="report.txt",
+            display_path_at_upload="records/report.txt",
+            size_bytes=1,
+            checksum_sha256="a" * 64,
+        )
+        self.session.add_all([self.blob, self.asset, self.version])
+        self.session.commit()
+
+    def tearDown(self) -> None:
+        self.session.close()
+        self.engine.dispose()
+
+    def test_preview_hash_changes_with_retention_and_blocks_active_policy(self) -> None:
+        first = preview_asset_purge(
+            self.session, tenant_id="tenant-1", file_ids=[self.asset.id]
+        )
+        self.assertTrue(first.eligible)
+
+        set_asset_lifecycle(
+            self.session,
+            self.asset,
+            retained_until=utcnow() + timedelta(days=7),
+            legal_hold=True,
+            reason="Active proceeding",
+            expected_revision=1,
+        )
+        self.session.commit()
+        second = preview_asset_purge(
+            self.session, tenant_id="tenant-1", file_ids=[self.asset.id]
+        )
+
+        self.assertFalse(second.eligible)
+        self.assertNotEqual(first.preview_sha256, second.preview_sha256)
+        self.assertEqual(
+            {"legal_hold", "retention_active"}, set(second.items[0].blockers)
+        )
+
+    def test_restore_preserves_version_and_provenance_and_rejects_collision(self) -> None:
+        self.asset.metadata_ = {"source": {"provider": "s3", "revision": "v1"}}
+        self.session.commit()
+
+        self.assertTrue(restore_asset(self.session, self.asset))
+        self.assertEqual("version-1", self.asset.current_version_id)
+        self.assertEqual("v1", self.asset.metadata_["source"]["revision"])
+        self.session.commit()
+
+        self.asset.deleted_at = utcnow()
+        collision = FileAsset(
+            id="file-2",
+            tenant_id="tenant-1",
+            owner_type="user",
+            owner_user_id="user-1",
+            display_path=self.asset.display_path,
+            filename=self.asset.filename,
+        )
+        self.session.add(collision)
+        self.session.commit()
+        with self.assertRaisesRegex(FileStorageError, "path is already in use"):
+            restore_asset(self.session, self.asset)
+
+    def test_active_share_blocks_purge_until_revoked(self) -> None:
+        share = FileShare(
+            id="share-1",
+            tenant_id="tenant-1",
+            file_asset_id=self.asset.id,
+            target_type="tenant",
+            target_id="tenant-1",
+            permission="read",
+        )
+        self.session.add(share)
+        self.session.commit()
+
+        blocked = preview_asset_purge(
+            self.session, tenant_id="tenant-1", file_ids=[self.asset.id]
+        )
+        self.assertIn("active_share", blocked.items[0].blockers)
+
+        share.revoked_at = utcnow()
+        self.session.commit()
+        allowed = preview_asset_purge(
+            self.session, tenant_id="tenant-1", file_ids=[self.asset.id]
+        )
+        self.assertTrue(allowed.eligible)
+
+    def test_missing_campaign_table_is_an_optional_module_safe_path(self) -> None:
+        preview = preview_asset_purge(
+            self.session, tenant_id="tenant-1", file_ids=[self.asset.id]
+        )
+        self.assertNotIn("campaign_evidence", preview.items[0].blockers)
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_form_evidence.py

@@ -0,0 +1,333 @@
+from __future__ import annotations
+
+from datetime import UTC, datetime, timedelta
+from io import BytesIO
+from types import SimpleNamespace
+import unittest
+from unittest.mock import patch
+
+from fastapi import UploadFile
+from sqlalchemy import create_engine
+from sqlalchemy.orm import Session
+from starlette.datastructures import Headers
+
+from govoplan_access.backend.db.models import Account, User
+from govoplan_core.core.access import (
+    CAPABILITY_ACCESS_DIRECTORY,
+    AccessSubjectRef,
+    UserRef,
+)
+from govoplan_core.core.change_sequence import ChangeSequenceEntry
+from govoplan_core.core.form_evidence import (
+    FormEvidenceContractError,
+    FormEvidenceGrantRequest,
+    FormEvidenceInspectionRequest,
+)
+from govoplan_core.core.institutional import EvidenceReference, InstitutionalReference
+from govoplan_core.db.base import Base
+from govoplan_files.backend.db.models import (
+    FileAsset,
+    FileBlob,
+    FileFormEvidenceGrant,
+    FileVersion,
+)
+from govoplan_files.backend.form_evidence import FilesFormEvidenceProvider
+from govoplan_files.backend.routes.form_evidence import upload_form_evidence
+
+
+class _Directory:
+    def __init__(self, user: UserRef) -> None:
+        self._user = user
+
+    def get_account(self, account_id: str):
+        del account_id
+        return None
+
+    def get_user(self, user_id: str):
+        return self._user if user_id == self._user.id else None
+
+    def get_users(self, user_ids):
+        return {user_id: self._user for user_id in user_ids if user_id == self._user.id}
+
+    def users_for_tenant(self, tenant_id: str):
+        return (self._user,) if tenant_id == self._user.tenant_id else ()
+
+    def get_group(self, group_id: str):
+        del group_id
+        return None
+
+    def get_groups(self, group_ids):
+        del group_ids
+        return {}
+
+    def groups_for_tenant(self, tenant_id: str):
+        del tenant_id
+        return ()
+
+    def groups_for_user(self, user_id: str, *, tenant_id: str):
+        del user_id, tenant_id
+        return ()
+
+    def display_label(self, subject: AccessSubjectRef):
+        del subject
+        return self._user.display_name
+
+
+class _Registry:
+    def __init__(self, directory: _Directory) -> None:
+        self._directory = directory
+
+    def has_capability(self, name: str) -> bool:
+        return name == CAPABILITY_ACCESS_DIRECTORY
+
+    def require_capability(self, name: str):
+        if not self.has_capability(name):
+            raise KeyError(name)
+        return self._directory
+
+
+class FilesFormEvidenceTests(unittest.TestCase):
+    def setUp(self) -> None:
+        self.engine = create_engine("sqlite+pysqlite:///:memory:")
+        Base.metadata.create_all(
+            self.engine,
+            tables=[
+                Account.__table__,
+                User.__table__,
+                FileBlob.__table__,
+                FileAsset.__table__,
+                FileVersion.__table__,
+                FileFormEvidenceGrant.__table__,
+                ChangeSequenceEntry.__table__,
+            ],
+        )
+        self.session = Session(self.engine)
+        user_ref = UserRef(
+            id="user-1",
+            account_id="account-1",
+            tenant_id="tenant-1",
+            email="user@example.test",
+            display_name="Evidence Custodian",
+        )
+        self.provider = FilesFormEvidenceProvider(
+            _Registry(_Directory(user_ref)),
+            SimpleNamespace(file_upload_max_bytes=2_000_000),
+        )
+        self.principal = SimpleNamespace(tenant_id="tenant-1")
+        self.definition_ref = InstitutionalReference(
+            kind="form",
+            owner_module="forms",
+            object_id="form-1",
+            tenant_id="tenant-1",
+            version="3",
+        )
+
+    def tearDown(self) -> None:
+        self.session.close()
+        self.engine.dispose()
+
+    def request(self, **overrides: object) -> FormEvidenceGrantRequest:
+        values: dict[str, object] = {
+            "tenant_id": "tenant-1",
+            "instance_id": "instance-1",
+            "definition_ref": self.definition_ref,
+            "evidence_kind": "document",
+            "purpose": "supporting document",
+            "idempotency_key": "grant-1",
+            "expires_at": datetime.now(UTC) + timedelta(hours=1),
+            "custodian_ref": "user:user-1",
+            "max_size_bytes": 1_000_000,
+            "allowed_content_types": ("application/pdf",),
+        }
+        values.update(overrides)
+        return FormEvidenceGrantRequest(**values)  # type: ignore[arg-type]
+
+    def test_grant_persists_only_token_hash_and_replays_without_secret(self) -> None:
+        request = self.request()
+        issued = self.provider.create_upload_grant(
+            self.session,
+            self.principal,
+            request=request,
+        )
+        row = self.session.query(FileFormEvidenceGrant).one()
+
+        self.assertIsNotNone(issued.upload_token)
+        self.assertNotEqual(issued.upload_token, row.token_sha256)
+        self.assertLessEqual(
+            issued.expires_at,
+            datetime.now(UTC) + timedelta(minutes=15, seconds=1),
+        )
+        replay = self.provider.create_upload_grant(
+            self.session,
+            self.principal,
+            request=request,
+        )
+        self.assertTrue(replay.replayed)
+        self.assertIsNone(replay.upload_token)
+        self.assertEqual(issued.grant_id, replay.grant_id)
+
+    def test_grant_capacity_counts_outstanding_uploads_but_allows_replay(self) -> None:
+        request = self.request(metadata={"remaining_attachments": 1})
+        self.provider.create_upload_grant(self.session, self.principal, request=request)
+
+        replay = self.provider.create_upload_grant(
+            self.session,
+            self.principal,
+            request=request,
+        )
+        self.assertTrue(replay.replayed)
+        with self.assertRaisesRegex(
+            FormEvidenceContractError,
+            "maximum number of active attachment uploads",
+        ):
+            self.provider.create_upload_grant(
+                self.session,
+                self.principal,
+                request=self.request(
+                    idempotency_key="grant-2",
+                    metadata={"remaining_attachments": 1},
+                ),
+            )
+
+    def test_idempotency_key_cannot_be_reused_for_another_request(self) -> None:
+        self.provider.create_upload_grant(
+            self.session,
+            self.principal,
+            request=self.request(),
+        )
+        with self.assertRaisesRegex(
+            FormEvidenceContractError,
+            "idempotency conflict",
+        ):
+            self.provider.create_upload_grant(
+                self.session,
+                self.principal,
+                request=self.request(purpose="another purpose"),
+            )
+
+    def test_inspection_accepts_only_exact_verified_managed_version(self) -> None:
+        issued = self.provider.create_upload_grant(
+            self.session,
+            self.principal,
+            request=self.request(),
+        )
+        grant = self.session.get(FileFormEvidenceGrant, issued.grant_id)
+        assert grant is not None
+        blob = FileBlob(
+            id="blob-1",
+            tenant_id="tenant-1",
+            storage_backend="local",
+            storage_key="tenant-1/blob-1",
+            checksum_sha256="a" * 64,
+            size_bytes=100,
+            integrity_status="verified",
+        )
+        asset = FileAsset(
+            id="asset-1",
+            tenant_id="tenant-1",
+            owner_type="user",
+            owner_user_id="user-1",
+            current_version_id="version-1",
+            display_path="Form submissions/instance-1/evidence.pdf",
+            filename="evidence.pdf",
+        )
+        version = FileVersion(
+            id="version-1",
+            tenant_id="tenant-1",
+            file_asset_id="asset-1",
+            blob_id="blob-1",
+            version_number=1,
+            filename_at_upload="evidence.pdf",
+            display_path_at_upload=asset.display_path,
+            content_type="application/pdf",
+            size_bytes=100,
+            checksum_sha256="a" * 64,
+        )
+        self.session.add_all((blob, asset, version))
+        grant.status = "uploaded"
+        grant.file_asset_id = asset.id
+        grant.file_version_id = version.id
+        self.session.flush()
+        reference = EvidenceReference(
+            kind="document",
+            owner_module="files",
+            evidence_id=asset.id,
+            tenant_id="tenant-1",
+            version=version.id,
+            checksum="a" * 64,
+        )
+
+        accepted = self.provider.inspect_evidence(
+            self.session,
+            self.principal,
+            request=FormEvidenceInspectionRequest(
+                tenant_id="tenant-1",
+                instance_id="instance-1",
+                definition_ref=self.definition_ref,
+                evidence=reference,
+                purpose="final submission",
+                final=True,
+            ),
+        )
+        self.assertTrue(accepted.accepted)
+
+        rejected = self.provider.inspect_evidence(
+            self.session,
+            self.principal,
+            request=FormEvidenceInspectionRequest(
+                tenant_id="tenant-1",
+                instance_id="another-instance",
+                definition_ref=self.definition_ref,
+                evidence=reference,
+                purpose="final submission",
+                final=True,
+            ),
+        )
+        self.assertEqual("rejected", rejected.state)
+
+    def test_public_upload_consumes_grant_and_returns_exact_evidence(self) -> None:
+        issued = self.provider.create_upload_grant(
+            self.session,
+            self.principal,
+            request=self.request(),
+        )
+        assert issued.upload_token is not None
+        stored = SimpleNamespace(
+            asset=SimpleNamespace(id="asset-uploaded"),
+            version=SimpleNamespace(
+                id="version-uploaded",
+                checksum_sha256="c" * 64,
+                size_bytes=8,
+            ),
+        )
+        upload = UploadFile(
+            filename="evidence.pdf",
+            file=BytesIO(b"evidence"),
+            headers=Headers({"content-type": "application/pdf"}),
+        )
+
+        with (
+            patch(
+                "govoplan_files.backend.routes.form_evidence.create_file_asset",
+                return_value=stored,
+            ) as create,
+            patch("govoplan_files.backend.routes.form_evidence.emit_platform_event"),
+        ):
+            response = upload_form_evidence(
+                file=upload,
+                x_form_evidence_token=issued.upload_token,
+                session=self.session,
+            )
+
+        grant = self.session.get(FileFormEvidenceGrant, issued.grant_id)
+        assert grant is not None
+        self.assertEqual("uploaded", grant.status)
+        self.assertEqual("asset-uploaded", grant.file_asset_id)
+        self.assertEqual("version-uploaded", grant.file_version_id)
+        self.assertEqual("asset-uploaded", response["evidence"]["evidence_id"])
+        self.assertEqual(b"evidence", create.call_args.kwargs["data"])
+        self.assertEqual("user-1", create.call_args.kwargs["owner_id"])
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_integrity_reconciliation.py

@@ -0,0 +1,211 @@
+from __future__ import annotations
+
+import hashlib
+import tempfile
+import unittest
+from pathlib import Path
+from unittest.mock import patch
+
+from fastapi import HTTPException
+from sqlalchemy import create_engine
+from sqlalchemy.orm import sessionmaker
+
+from govoplan_access.backend.db.models import Account, Group, User
+from govoplan_core.db.base import Base
+from govoplan_files.backend.db.models import (
+    FileBlob,
+    FileIntegrityFinding,
+    FileIntegrityScan,
+)
+from govoplan_files.backend.routes.integrity import _assert_expected_revision
+from govoplan_files.backend.storage.backends import LocalFilesystemStorageBackend
+from govoplan_files.backend.storage.common import FileStorageError
+from govoplan_files.backend.storage.integrity import (
+    cleanup_orphan_finding,
+    create_integrity_scan,
+    read_verified_blob_bytes,
+    recheck_integrity_finding,
+    run_integrity_scan_batch,
+)
+
+
+TENANT_ID = "tenant-1"
+USER_ID = "user-1"
+
+
+class IntegrityReconciliationTests(unittest.TestCase):
+    def setUp(self) -> None:
+        self.temporary_directory = tempfile.TemporaryDirectory()
+        self.addCleanup(self.temporary_directory.cleanup)
+        self.backend = LocalFilesystemStorageBackend(
+            Path(self.temporary_directory.name)
+        )
+        self.engine = create_engine("sqlite:///:memory:", future=True)
+        Base.metadata.create_all(
+            bind=self.engine,
+            tables=[
+                Account.__table__,
+                User.__table__,
+                Group.__table__,
+                FileBlob.__table__,
+                FileIntegrityScan.__table__,
+                FileIntegrityFinding.__table__,
+            ],
+        )
+        self.session = sessionmaker(bind=self.engine, future=True)()
+        self.addCleanup(self._close)
+
+    def _close(self) -> None:
+        self.session.close()
+        self.engine.dispose()
+
+    def test_scan_is_bounded_resumable_and_reconciles_both_orphan_directions(
+        self,
+    ) -> None:
+        valid_data = b"valid"
+        restored_data = b"restore-me"
+        expected_corrupt_data = b"expected"
+        stored_corrupt_data = b"corrupt!"
+        valid = _blob("blob-1", "valid.bin", valid_data)
+        missing = _blob("blob-2", "missing.bin", restored_data)
+        corrupt = _blob("blob-3", "corrupt.bin", expected_corrupt_data)
+        self.session.add_all([valid, missing, corrupt])
+        self.session.commit()
+        self.backend.put_bytes(valid.storage_key, valid_data)
+        self.backend.put_bytes(corrupt.storage_key, stored_corrupt_data)
+        orphan_key = f"tenants/{TENANT_ID}/files/orphan.bin"
+        self.backend.put_bytes(orphan_key, b"orphan")
+
+        scan = create_integrity_scan(
+            self.session,
+            tenant_id=TENANT_ID,
+            user_id=USER_ID,
+            batch_size=1,
+            backend=self.backend,
+        )
+        self.session.commit()
+        self.assertEqual(1, scan.revision)
+
+        invocations = 0
+        while scan.status != "completed":
+            run_integrity_scan_batch(
+                self.session,
+                scan,
+                backend=self.backend,
+            )
+            self.session.commit()
+            invocations += 1
+            scan = self.session.get(FileIntegrityScan, scan.id)
+            self.assertIsNotNone(scan)
+            self.session.expire_all()
+            if invocations > 12:
+                self.fail("Integrity scan did not complete")
+
+        self.assertGreater(invocations, 3)
+        self.assertEqual(1 + invocations, scan.revision)
+        self.assertEqual(3, scan.scanned_blob_count)
+        self.assertEqual(1, scan.verified_blob_count)
+        self.assertEqual(2, scan.quarantined_blob_count)
+        self.assertEqual(3, scan.scanned_object_count)
+        self.assertEqual(1, scan.orphan_object_count)
+        findings = (
+            self.session.query(FileIntegrityFinding)
+            .filter(FileIntegrityFinding.scan_id == scan.id)
+            .all()
+        )
+        self.assertEqual(
+            {"missing", "checksum_mismatch", "orphan_object"},
+            {finding.kind for finding in findings},
+        )
+
+        missing = self.session.get(FileBlob, missing.id)
+        corrupt = self.session.get(FileBlob, corrupt.id)
+        self.assertEqual("missing", missing.integrity_status)
+        self.assertEqual("checksum_mismatch", corrupt.integrity_status)
+        with self.assertRaisesRegex(FileStorageError, "quarantined"):
+            read_verified_blob_bytes(missing, backend=self.backend)
+
+        missing_finding = next(
+            finding for finding in findings if finding.kind == "missing"
+        )
+        self.backend.put_bytes(missing.storage_key, restored_data)
+        preview = recheck_integrity_finding(
+            self.session,
+            missing_finding,
+            user_id=USER_ID,
+            dry_run=True,
+            backend=self.backend,
+        )
+        self.assertTrue(preview.inspection.valid)
+        self.assertEqual("open", missing_finding.state)
+        repaired = recheck_integrity_finding(
+            self.session,
+            missing_finding,
+            user_id=USER_ID,
+            dry_run=False,
+            backend=self.backend,
+        )
+        self.session.commit()
+        self.assertTrue(repaired.changed)
+        self.assertEqual("resolved", missing_finding.state)
+        self.assertEqual(
+            restored_data,
+            read_verified_blob_bytes(missing, backend=self.backend),
+        )
+
+        orphan_finding = next(
+            finding for finding in findings if finding.kind == "orphan_object"
+        )
+        preview_cleanup = cleanup_orphan_finding(
+            self.session,
+            orphan_finding,
+            user_id=USER_ID,
+            dry_run=True,
+            backend=self.backend,
+        )
+        self.assertEqual("would_delete", preview_cleanup.action)
+        self.assertTrue(self.backend.exists(orphan_key))
+        with patch(
+            "govoplan_files.backend.storage.integrity.begin_orphan_cleanup_recovery"
+        ):
+            cleanup = cleanup_orphan_finding(
+                self.session,
+                orphan_finding,
+                user_id=USER_ID,
+                dry_run=False,
+                backend=self.backend,
+            )
+            repeated = cleanup_orphan_finding(
+                self.session,
+                orphan_finding,
+                user_id=USER_ID,
+                dry_run=False,
+                backend=self.backend,
+            )
+        self.assertEqual("deleted", cleanup.action)
+        self.assertFalse(self.backend.exists(orphan_key))
+        self.assertEqual("already_deleted", repeated.action)
+        self.assertFalse(repeated.changed)
+
+    def test_stale_integrity_action_revision_is_rejected(self) -> None:
+        with self.assertRaises(HTTPException) as captured:
+            _assert_expected_revision(3, 2)
+
+        self.assertEqual(409, captured.exception.status_code)
+        self.assertIn("reload", str(captured.exception.detail).lower())
+
+
+def _blob(blob_id: str, filename: str, expected_data: bytes) -> FileBlob:
+    return FileBlob(
+        id=blob_id,
+        tenant_id=TENANT_ID,
+        storage_backend="local",
+        storage_key=f"tenants/{TENANT_ID}/files/{filename}",
+        checksum_sha256=hashlib.sha256(expected_data).hexdigest(),
+        size_bytes=len(expected_data),
+        ref_count=1,
+    )
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_manifest_documentation.py

@@ -4,13 +4,24 @@ import unittest
 
 
 STATIC_TOPIC_IDS = {
+    "files.configuration-package.managed-storage",
+    "files.quick-access-and-product-area",
+    "files.search.managed-content",
     "files.workflow.organize-managed-files",
     "files.workflow.find-and-download-files",
     "files.workflow.share-managed-files",
     "files.workflow.delete-managed-files",
+    "files.workflow.restore-retain-and-purge",
+    "files.privacy.data-subject-requests",
     "files.governed-connectors-and-provenance",
     "files.reference.integrity-recovery-and-fail-closed-transports",
+    "files.reference.shared-storage-profile",
+    "files.reference.generated-artifact-store",
     "files.reference.snapshot-provenance-and-capabilities",
+    "files.records.exact-version-source",
+    "files.forms-runtime.managed-evidence",
+    "files.postbox.exact-version-references",
+    "files.tabular-content",
     "files.assurance.process-and-release-readiness",
 }
 RUNTIME_TOPIC_IDS = {
@@ -103,17 +114,30 @@ class FilesManifestDocumentationTests(unittest.TestCase):
     def test_share_and_delete_tasks_state_current_boundaries(self) -> None:
         share = self.topic("files.workflow.share-managed-files")
         self.assertEqual("available", share.layer)
-        self.assertIn("does not yet provide a general share editor", share.body)
+        self.assertIn("Expired and revoked grants stop authorizing", share.body)
-        self.assertIn("no share-revocation route", share.body)
+        self.assertIn("revocation is idempotent", share.body)
         self.assertIn(
             "/api/v1/files/{file_id}/shares", {link.href for link in share.links}
         )
 
         delete = self.topic("files.workflow.delete-managed-files")
         self.assertIn("Soft-delete", delete.summary)
-        self.assertIn("no self-service restore or hard-purge", delete.body)
+        self.assertIn("Authorized restoration", delete.body)
         self.assertTrue(
-            any("not a hard purge" in item for item in delete.metadata["limitations"])
+            any("soft deletion" in item.casefold() for item in delete.metadata["limitations"])
+        )
+
+        lifecycle = self.topic("files.workflow.restore-retain-and-purge")
+        self.assertIn("legal hold", lifecycle.body)
+        self.assertIn("preview hash", lifecycle.body)
+        self.assertIn("recovery ledger", lifecycle.body)
+
+        privacy = self.topic("files.privacy.data-subject-requests")
+        self.assertEqual(("admin",), privacy.documentation_types)
+        self.assertIn("without raw file bytes", privacy.body)
+        self.assertIn("separately authorized Files purge", privacy.body)
+        self.assertTrue(
+            any("membership" in item for item in privacy.metadata["limitations"])
         )
 
     def test_admin_topic_covers_policy_redaction_and_atomic_credential_deletion(
@@ -125,9 +149,28 @@ class FilesManifestDocumentationTests(unittest.TestCase):
         self.assertEqual("reference", topic.metadata["kind"])
         self.assertEqual("File connections", topic.metadata["screen"])
         self.assertTrue(topic.metadata["section"])
+        self.assertEqual(
+            {
+                "files.connectors",
+                "files.connector.credentials",
+                "files.connector.policy",
+                "files.connector-folder-sync.remote-path",
+                "files.connector-folder-sync.target-folder",
+                "files.connector-folder-sync.conflict-strategy",
+                "files.connector-folder-sync.max-files",
+                "files.connector-folder-sync.recursive",
+            },
+            set(topic.metadata["help_contexts"]),
+        )
         self.assertIn("deny rules win", topic.body)
         self.assertIn("redact secret values", topic.body)
         self.assertIn("same transaction", topic.body)
+        self.assertIn("read-only by default", topic.body)
+        self.assertIn("S3", topic.body)
+        self.assertIn("manual connector-folder sync", topic.body)
+        self.assertIn("Per-file savepoints", topic.body)
+        self.assertIn("files.connector.folder_synced", topic.body)
+        self.assertIn("Automatic remote deletion", topic.body)
         self.assertTrue(
             any(
                 "non-owned external references" in item
@@ -141,8 +184,13 @@ class FilesManifestDocumentationTests(unittest.TestCase):
         self.assertIn(
             "/api/v1/files/connectors/credentials", {link.href for link in topic.links}
         )
+        self.assertIn(
+            "/api/v1/files/connector-spaces/{space_id}/sync",
+            {link.href for link in topic.links},
+        )
+        self.assertIn("de", topic.translations)
 
-    def test_operator_topic_covers_recovery_and_fail_closed_s3_smb(self) -> None:
+    def test_operator_topic_covers_recovery_and_pinned_s3_smb(self) -> None:
         topic = self.topic(
             "files.reference.integrity-recovery-and-fail-closed-transports"
         )
@@ -153,9 +201,31 @@ class FilesManifestDocumentationTests(unittest.TestCase):
         self.assertIn("SMB", topic.body)
         self.assertIn("fail closed", topic.body)
         self.assertIn("DFS referrals", topic.body)
+        self.assertIn("ambient credential discovery", topic.body)
         self.assertIn("does not remove backend blob objects", topic.body)
+        self.assertIn("bounded resumable integrity scan", topic.body)
+        self.assertIn("quarantined", topic.body)
         self.assertIn("MASTER_KEY_B64", topic.metadata["recovery_unit"])
+        self.assertIn(
+            "Encryption envelope and wrapped-key rows", topic.metadata["recovery_unit"]
+        )
+        self.assertIn("lease-fenced Core recovery", topic.body)
+        self.assertIn("Ops", topic.body)
+        self.assertIn("Hard purge", topic.body)
+        self.assertIn("S3 connector write-back", topic.body)
         self.assertTrue(topic.metadata["verification"])
+        self.assertIn(
+            "/api/v1/files/integrity/scans",
+            {link.href for link in topic.links},
+        )
+        self.assertIn(
+            "/admin?section=tenant-file-integrity",
+            {link.href for link in topic.links},
+        )
+        self.assertIn(
+            "files.admin.tenant-integrity",
+            {surface.id for surface in self.manifest.frontend.view_surfaces},
+        )
         self.assertIn(
             "GOVOPLAN_CONNECTOR_ALLOW_PRIVATE_NETWORKS", topic.configuration_keys
         )
@@ -166,7 +236,11 @@ class FilesManifestDocumentationTests(unittest.TestCase):
         self.assertEqual(("admin", "user"), topic.documentation_types)
         self.assertEqual("reference", topic.metadata["kind"])
         self.assertEqual(
-            ["files.access@0.1.6", "files.campaign_attachments@0.1.6"],
+            [
+                "files.access@0.1.6",
+                "files.campaign_attachments@0.1.6",
+                "forms_runtime.evidence.files@1.0.0",
+            ],
             topic.metadata["provided_interfaces"],
         )
         self.assertIn("revision", topic.metadata["provenance_fields"])
@@ -184,8 +258,9 @@ class FilesManifestDocumentationTests(unittest.TestCase):
         self.assertIn("process_owner", topic.audience)
         self.assertIn("release_manager", topic.audience)
         self.assertIn("versions align", topic.body)
-        self.assertIn("no general share-management UI or share revocation", topic.body)
+        self.assertIn("share lifecycle", topic.body)
-        self.assertIn("no enforced retention or legal hold", topic.body)
+        self.assertIn("legal hold", topic.body)
+        self.assertIn("S3 write-back", topic.body)
         for key in ("prerequisites", "steps", "outcome", "verification"):
             self.assertTrue(topic.metadata[key])
         self.assertTrue(

tests/test_migrations.py

@@ -0,0 +1,42 @@
+from __future__ import annotations
+
+from pathlib import Path
+import tempfile
+import unittest
+
+from alembic.runtime.migration import MigrationContext
+from sqlalchemy import create_engine, inspect
+
+from govoplan_access.backend.manifest import get_manifest as get_access_manifest
+from govoplan_core.db.migrations import migrate_database
+from govoplan_files.backend.manifest import get_manifest as get_files_manifest
+
+
+class FilesMigrationTests(unittest.TestCase):
+    def test_fresh_migration_creates_form_evidence_grants_and_head(self) -> None:
+        with tempfile.TemporaryDirectory(
+            prefix="govoplan-files-migration-"
+        ) as directory:
+            url = f"sqlite:///{Path(directory) / 'files.db'}"
+            migrate_database(
+                database_url=url,
+                enabled_modules=("access", "files"),
+                manifest_factories=(get_access_manifest, get_files_manifest),
+            )
+            engine = create_engine(url)
+            try:
+                self.assertIn(
+                    "file_form_evidence_grants",
+                    inspect(engine).get_table_names(),
+                )
+                with engine.connect() as connection:
+                    self.assertIn(
+                        "a2b3c4d5e6f9",
+                        set(MigrationContext.configure(connection).get_current_heads()),
+                    )
+            finally:
+                engine.dispose()
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_operational_checks.py

@@ -0,0 +1,42 @@
+from __future__ import annotations
+
+from unittest.mock import patch
+
+from govoplan_files.backend.operational_checks import managed_storage_roundtrip_check
+from govoplan_files.backend.storage.backends import LocalFilesystemStorageBackend
+
+
+def test_managed_storage_roundtrip_removes_probe(tmp_path) -> None:
+    backend = LocalFilesystemStorageBackend(tmp_path)
+
+    with patch(
+        "govoplan_files.backend.operational_checks.get_storage_backend",
+        return_value=backend,
+    ):
+        result = managed_storage_roundtrip_check()
+
+    assert result.state == "ok"
+    assert result.metrics["backend"] == "local"
+    assert list(tmp_path.rglob("*.bin")) == []
+
+
+def test_managed_storage_roundtrip_fails_closed() -> None:
+    class BrokenBackend:
+        name = "broken"
+
+        def put_bytes(self, *_args, **_kwargs):
+            raise OSError("private provider detail")
+
+        def delete(self, *_args, **_kwargs):
+            return None
+
+    with patch(
+        "govoplan_files.backend.operational_checks.get_storage_backend",
+        return_value=BrokenBackend(),
+    ):
+        result = managed_storage_roundtrip_check()
+
+    assert result.state == "error"
+    assert result.readiness_critical is True
+    assert "private provider detail" not in result.detail
+

tests/test_postbox_reference.py

@@ -0,0 +1,160 @@
+from __future__ import annotations
+
+from types import SimpleNamespace
+import unittest
+from unittest.mock import patch
+
+from govoplan_core.core.files import PostboxFileReferenceRequest
+from govoplan_files.backend.capabilities import FilesPostboxReferenceService
+from govoplan_files.backend.db.models import FileAsset, FileBlob, FileVersion
+from govoplan_files.backend.storage.common import FileStorageError
+
+
+class _Principal:
+    def __init__(self, *, tenant_id: str = "tenant-1", scopes: set[str] | None = None):
+        self.tenant_id = tenant_id
+        self.user = SimpleNamespace(id="user-1")
+        self._scopes = {"files:file:download"} if scopes is None else scopes
+
+    def has(self, scope: str) -> bool:
+        return scope in self._scopes
+
+
+class _Session:
+    def __init__(self, values: dict[tuple[type[object], str], object]):
+        self.values = values
+
+    def get(self, model: type[object], resource_id: str):
+        return self.values.get((model, resource_id))
+
+
+def _request(reference_type: str = "file_version", reference_id: str = "version-1"):
+    return PostboxFileReferenceRequest(
+        postbox_id="postbox-1",
+        message_id="message-1",
+        reference_type=reference_type,
+        reference_id=reference_id,
+    )
+
+
+def _objects():
+    asset = SimpleNamespace(
+        id="file-1",
+        tenant_id="tenant-1",
+        current_version_id="version-2",
+        deleted_at=None,
+        filename="current.txt",
+        display_path="Evidence/current.txt",
+    )
+    version = SimpleNamespace(
+        id="version-1",
+        tenant_id="tenant-1",
+        file_asset_id="file-1",
+        blob_id="blob-1",
+        filename_at_upload="evidence.txt",
+        content_type="text/plain",
+        size_bytes=8,
+        checksum_sha256="a" * 64,
+    )
+    blob = SimpleNamespace(
+        id="blob-1",
+        tenant_id="tenant-1",
+        content_type="text/plain",
+    )
+    return asset, version, blob
+
+
+class FilesPostboxReferenceTests(unittest.TestCase):
+    def test_resolves_the_referenced_version_instead_of_the_current_version(self) -> None:
+        asset, version, blob = _objects()
+        session = _Session(
+            {
+                (FileAsset, asset.id): asset,
+                (FileVersion, version.id): version,
+                (FileBlob, blob.id): blob,
+            }
+        )
+
+        with patch(
+            "govoplan_files.backend.capabilities.get_asset_for_user",
+            return_value=asset,
+        ):
+            result = FilesPostboxReferenceService().resolve_postbox_references(
+                session,
+                _Principal(),
+                tenant_id="tenant-1",
+                requests=(_request(),),
+            )[0]
+
+        self.assertTrue(result.available)
+        self.assertEqual("version-1", result.file_version_id)
+        self.assertEqual(
+            "/api/v1/files/file-1/versions/version-1/download",
+            result.download_path,
+        )
+        self.assertTrue(result.provenance["exact_version"])
+
+    def test_fails_closed_for_tenant_permission_and_file_access_mismatches(self) -> None:
+        service = FilesPostboxReferenceService()
+        empty = _Session({})
+
+        tenant_result = service.resolve_postbox_references(
+            empty,
+            _Principal(tenant_id="tenant-2"),
+            tenant_id="tenant-1",
+            requests=(_request(),),
+        )[0]
+        permission_result = service.resolve_postbox_references(
+            empty,
+            _Principal(scopes=set()),
+            tenant_id="tenant-1",
+            requests=(_request(),),
+        )[0]
+
+        asset, version, blob = _objects()
+        populated = _Session(
+            {
+                (FileAsset, asset.id): asset,
+                (FileVersion, version.id): version,
+                (FileBlob, blob.id): blob,
+            }
+        )
+        with patch(
+            "govoplan_files.backend.capabilities.get_asset_for_user",
+            side_effect=FileStorageError("denied"),
+        ):
+            access_result = service.resolve_postbox_references(
+                populated,
+                _Principal(),
+                tenant_id="tenant-1",
+                requests=(_request(),),
+            )[0]
+
+        self.assertEqual("tenant_mismatch", tenant_result.reason_code)
+        self.assertEqual("download_permission_missing", permission_result.reason_code)
+        self.assertEqual("file_access_denied", access_result.reason_code)
+        self.assertFalse(access_result.available)
+
+    def test_asset_reference_does_not_drift_to_a_later_current_version(self) -> None:
+        asset, version, blob = _objects()
+        session = _Session(
+            {
+                (FileAsset, asset.id): asset,
+                (FileVersion, version.id): version,
+                (FileBlob, blob.id): blob,
+            }
+        )
+
+        result = FilesPostboxReferenceService().resolve_postbox_references(
+            session,
+            _Principal(),
+            tenant_id="tenant-1",
+            requests=(_request("file", asset.id),),
+        )[0]
+
+        self.assertFalse(result.available)
+        self.assertEqual("exact_version_required", result.reason_code)
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_provider_state.py

@@ -0,0 +1,97 @@
+from __future__ import annotations
+
+import unittest
+
+from sqlalchemy import create_engine
+from sqlalchemy.orm import sessionmaker
+
+from govoplan_access.backend.db import models as access_models  # noqa: F401
+from govoplan_core.core.provider_governance import ExternalProviderStateContext
+from govoplan_core.db.base import Base
+from govoplan_files.backend.db.models import FileConnectorProfile, FileConnectorSpace
+from govoplan_files.backend.manifest import manifest
+from govoplan_files.backend.provider_state import (
+    REMOTE_STORAGE_PROVIDER_ID,
+    remote_storage_provider_states,
+)
+
+
+class FilesProviderStateTests(unittest.TestCase):
+    def setUp(self) -> None:
+        self.engine = create_engine("sqlite+pysqlite:///:memory:", future=True)
+        Base.metadata.create_all(
+            self.engine,
+            tables=(FileConnectorProfile.__table__, FileConnectorSpace.__table__),
+        )
+        self.session = sessionmaker(bind=self.engine, expire_on_commit=False)()
+        self.profile = FileConnectorProfile(
+            id="profile-1",
+            tenant_id="tenant-1",
+            scope_type="tenant",
+            scope_id="tenant-1",
+            label="Remote WebDAV",
+            provider="webdav",
+            endpoint_url="https://files.example.test/dav/",
+            enabled=True,
+        )
+        self.session.add_all(
+            (
+                self.profile,
+                FileConnectorSpace(
+                    id="space-1",
+                    tenant_id="tenant-1",
+                    owner_type="tenant",
+                    label="Documents",
+                    connector_profile_id=self.profile.id,
+                    provider="webdav",
+                    remote_path="/documents",
+                    read_only=True,
+                    is_active=True,
+                ),
+            )
+        )
+        self.session.commit()
+
+    def tearDown(self) -> None:
+        self.session.close()
+        self.engine.dispose()
+
+    def test_state_does_not_overstate_live_health_or_expose_endpoint(self) -> None:
+        state = remote_storage_provider_states(
+            ExternalProviderStateContext(session=self.session, tenant_id="tenant-1")
+        )[0]
+
+        self.assertEqual("external_mirror", state.authority_mode)
+        self.assertEqual("unknown", state.health)
+        self.assertEqual("attention", state.recovery)
+        self.assertEqual(1, state.metrics["active_spaces"])
+        self.assertNotIn("files.example.test", str(state.to_dict()))
+        self.assertEqual(
+            (),
+            remote_storage_provider_states(
+                ExternalProviderStateContext(
+                    session=self.session,
+                    tenant_id="tenant-2",
+                )
+            ),
+        )
+
+    def test_unimplemented_profile_is_fail_closed_and_manifest_registered(self) -> None:
+        self.profile.provider = "sharepoint"
+        self.session.flush()
+
+        state = remote_storage_provider_states(
+            ExternalProviderStateContext(session=self.session, tenant_id="tenant-1")
+        )[0]
+
+        self.assertEqual("error", state.health)
+        self.assertEqual("unsupported", state.recovery)
+        self.assertEqual(REMOTE_STORAGE_PROVIDER_ID, manifest.external_providers[0].id)
+        self.assertEqual(
+            REMOTE_STORAGE_PROVIDER_ID,
+            manifest.external_provider_state_providers[0].provider_id,
+        )
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_quick_access_contract.py

@@ -0,0 +1,38 @@
+from __future__ import annotations
+
+from pathlib import Path
+import unittest
+
+from govoplan_files.backend.manifest import manifest
+
+
+REPOSITORY_ROOT = Path(__file__).resolve().parents[1]
+
+
+class FilesQuickAccessContractTests(unittest.TestCase):
+    def test_manifest_declares_exact_version_result_and_fallback(self) -> None:
+        tool = manifest.frontend.quick_access_tools[0]
+
+        self.assertEqual("files.recent", tool.id)
+        self.assertEqual(("files.file-version",), tool.returned_reference_kinds)
+        self.assertEqual("files.quick_access.files", tool.help_context_id)
+        self.assertEqual("/files", tool.full_page_path)
+
+    def test_renderer_and_page_keep_recent_selection_and_upload_bounded(self) -> None:
+        renderer = (
+            REPOSITORY_ROOT / "webui/src/features/files/FileQuickAccess.tsx"
+        ).read_text(encoding="utf-8")
+        page = (
+            REPOSITORY_ROOT / "webui/src/features/files/FilesPage.tsx"
+        ).read_text(encoding="utf-8")
+
+        self.assertIn('sort: "recent", page_size: 7', renderer)
+        self.assertIn('kind: "file-version"', renderer)
+        self.assertIn("quickAccessLaunchState(launchContext)", renderer)
+        self.assertIn('to="/files?quickAction=upload"', renderer)
+        self.assertIn('parameters.get("quickAction") !== "upload"', page)
+        self.assertIn('openDialog("upload"', page)
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_record_source.py

@@ -0,0 +1,108 @@
+from __future__ import annotations
+
+from dataclasses import dataclass, field
+from datetime import UTC, datetime
+from types import SimpleNamespace
+import unittest
+from unittest.mock import MagicMock, patch
+
+from sqlalchemy.orm import Session
+
+from govoplan_core.core.records import RecordContractError, RecordSourceLocator
+from govoplan_files.backend.db.models import FileAsset, FileBlob, FileVersion
+from govoplan_files.backend.record_source import FilesRecordSource
+
+
+@dataclass
+class Principal:
+    tenant_id: str = "tenant-1"
+    membership_id: str = "user-1"
+    user: object = field(default_factory=lambda: SimpleNamespace(id="user-1"))
+
+    def has(self, scope: str) -> bool:
+        return scope in {"files:file:read"}
+
+
+class FilesRecordSourceTests(unittest.TestCase):
+    def setUp(self) -> None:
+        self.session = MagicMock(spec=Session)
+        self.asset = FileAsset(
+            id="asset-1",
+            tenant_id="tenant-1",
+            owner_type="user",
+            owner_user_id="user-1",
+            current_version_id="version-1",
+            display_path="Evidence/Decision.pdf",
+            filename="Decision.pdf",
+        )
+        self.blob = FileBlob(
+            id="blob-1",
+            tenant_id="tenant-1",
+            storage_backend="local",
+            storage_key="tenant-1/blob-1",
+            checksum_sha256="a" * 64,
+            size_bytes=1024,
+            integrity_status="verified",
+        )
+        self.version = FileVersion(
+            id="version-1",
+            tenant_id="tenant-1",
+            file_asset_id="asset-1",
+            blob_id="blob-1",
+            version_number=1,
+            filename_at_upload="Decision.pdf",
+            display_path_at_upload="Evidence/Decision.pdf",
+            content_type="application/pdf",
+            size_bytes=1024,
+            checksum_sha256="a" * 64,
+        )
+        self.version.created_at = datetime.now(UTC)
+        version_query = self.session.query.return_value.filter.return_value
+        version_query.filter.return_value.one_or_none.return_value = self.version
+        self.session.get.return_value = self.blob
+
+    def locator(self) -> RecordSourceLocator:
+        return RecordSourceLocator(
+            tenant_id="tenant-1",
+            source_module="files",
+            resource_type="file_version",
+            resource_id="asset-1",
+            source_revision="version-1",
+        )
+
+    def test_resolves_exact_authorized_version_and_digest(self) -> None:
+        with patch(
+            "govoplan_files.backend.record_source.get_asset_for_user",
+            return_value=self.asset,
+        ):
+            result = FilesRecordSource().resolve(
+                self.session,
+                Principal(),
+                locator=self.locator(),
+                purpose="document decision basis",
+            )
+
+        self.assertEqual("version-1", result.locator.source_revision)
+        self.assertEqual("a" * 64, result.content_sha256)
+        self.assertEqual("verified", result.metadata["integrity_status"])
+
+    def test_quarantined_version_fails_closed(self) -> None:
+        self.blob.quarantined_at = datetime.now(UTC)
+        self.session.flush()
+        with (
+            patch(
+                "govoplan_files.backend.record_source.get_asset_for_user",
+                return_value=self.asset,
+            ),
+            self.assertRaisesRegex(RecordContractError, "integrity gate"),
+        ):
+            FilesRecordSource().resolve(
+                self.session,
+                Principal(),
+                locator=self.locator(),
+                purpose="document decision basis",
+            )
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_router_contract.py

@@ -1,13 +1,79 @@
 from __future__ import annotations
 
 import unittest
+from collections import Counter
+from inspect import signature
 
 from govoplan_files.backend.router import router
+from govoplan_files.backend.routes.assets import router as assets_router
+from govoplan_files.backend.routes.connector_io import router as connector_io_router
+from govoplan_files.backend.routes.connector_profiles import (
+    router as connector_profiles_router,
+)
+from govoplan_files.backend.routes.connector_settings import (
+    router as connector_settings_router,
+)
+from govoplan_files.backend.routes.folders import router as folders_router
+from govoplan_files.backend.routes.form_evidence import router as form_evidence_router
+from govoplan_files.backend.routes.integrity import router as integrity_router
+from govoplan_files.backend.routes.listing import router as listing_router
+from govoplan_files.backend.routes.lifecycle import router as lifecycle_router
+from govoplan_files.backend.routes.shares import router as shares_router
+from govoplan_files.backend.routes.spaces import router as spaces_router
+from govoplan_files.backend.routes.transfers import router as transfers_router
+from govoplan_files.backend.routes.uploads import router as uploads_router
 
 
 class FilesRouterContractTests(unittest.TestCase):
+    @staticmethod
+    def _operation_keys(candidate_router) -> list[tuple[str, str]]:
+        return [
+            (method, route.path)
+            for route in candidate_router.routes
+            for method in sorted(route.methods or ())
+        ]
+
+    def test_composed_router_contains_every_workflow_operation_once(self) -> None:
+        workflow_routers = (
+            spaces_router,
+            folders_router,
+            form_evidence_router,
+            integrity_router,
+            lifecycle_router,
+            listing_router,
+            uploads_router,
+            connector_settings_router,
+            connector_io_router,
+            connector_profiles_router,
+            assets_router,
+            shares_router,
+            transfers_router,
+        )
+        expected = [
+            operation
+            for workflow_router in workflow_routers
+            for operation in self._operation_keys(workflow_router)
+        ]
+        actual = self._operation_keys(router)
+
+        self.assertEqual(expected, actual)
+        self.assertEqual(63, len(actual))
+        self.assertFalse(
+            [operation for operation, count in Counter(actual).items() if count > 1]
+        )
+
+    def test_archive_preview_and_confirmation_routes_are_exposed(self) -> None:
+        routes = {
+            (tuple(sorted(route.methods or ())), route.path) for route in router.routes
+        }
+
+        self.assertIn((("POST",), "/files/archive-preview"), routes)
+        self.assertIn((("POST",), "/files/archive-confirm"), routes)
+
     def test_connector_routes_keep_existing_api_paths(self) -> None:
-        routes = {(tuple(sorted(route.methods or ())), route.path) for route in router.routes}
+        routes = {
+            (tuple(sorted(route.methods or ())), route.path) for route in router.routes
+        }
 
         expected = {
             (("GET",), "/files/connectors/providers"),
@@ -17,20 +83,70 @@ class FilesRouterContractTests(unittest.TestCase):
             (("GET",), "/files/connectors/profiles/{profile_id}/browse"),
             (("POST",), "/files/connectors/profiles/{profile_id}/import"),
             (("POST",), "/files/connectors/profiles/{profile_id}/sync"),
+            (("POST",), "/files/connector-spaces/{space_id}/sync"),
+            (("POST",), "/files/connector-spaces/{space_id}/write-back"),
             (("GET",), "/files/connectors/credentials"),
             (("POST",), "/files/connectors/credentials"),
             (("GET",), "/files/connector-spaces"),
             (("POST",), "/files/connector-spaces"),
+            (("PATCH",), "/files/connector-spaces/{space_id}"),
+            (("DELETE",), "/files/connector-spaces/{space_id}"),
         }
 
         self.assertTrue(expected.issubset(routes))
 
     def test_bulk_organize_routes_keep_existing_api_paths(self) -> None:
-        routes = {(tuple(sorted(route.methods or ())), route.path) for route in router.routes}
+        routes = {
+            (tuple(sorted(route.methods or ())), route.path) for route in router.routes
+        }
 
         self.assertIn((("POST",), "/files/bulk-rename"), routes)
         self.assertIn((("POST",), "/files/transfer"), routes)
 
+    def test_share_lifecycle_routes_are_exposed(self) -> None:
+        routes = {
+            (tuple(sorted(route.methods or ())), route.path) for route in router.routes
+        }
+
+        self.assertIn((("GET",), "/files/{file_id}/shares"), routes)
+        self.assertIn((("POST",), "/files/{file_id}/shares"), routes)
+        self.assertIn((("DELETE",), "/files/{file_id}/shares/{share_id}"), routes)
+        self.assertIn((("GET",), "/files/{file_id}/share-target-options"), routes)
+
+    def test_exact_version_download_route_is_exposed(self) -> None:
+        routes = {
+            (tuple(sorted(route.methods or ())), route.path) for route in router.routes
+        }
+
+        self.assertIn(
+            (("GET",), "/files/{file_id}/versions/{version_id}/download"),
+            routes,
+        )
+
+    def test_governed_lifecycle_routes_are_exposed(self) -> None:
+        routes = {
+            (tuple(sorted(route.methods or ())), route.path) for route in router.routes
+        }
+
+        self.assertIn((("PATCH",), "/files/{file_id}/lifecycle"), routes)
+        self.assertIn((("POST",), "/files/assets/{file_id}/restore"), routes)
+        self.assertIn((("POST",), "/files/purge/preview"), routes)
+        self.assertIn((("POST",), "/files/purge/execute"), routes)
+        self.assertIn((("POST",), "/files/purge/blobs"), routes)
+
+    def test_file_listing_exposes_structured_property_filters(self) -> None:
+        route = next(
+            route
+            for route in listing_router.routes
+            if route.path == "/files" and "GET" in (route.methods or ())
+        )
+        parameters = signature(route.endpoint).parameters
+
+        self.assertIn("campaign_usage", parameters)
+        self.assertIn("audit_relevant", parameters)
+        self.assertIn("sort", parameters)
+        self.assertIn("total", route.response_model.model_fields)
+
 
 if __name__ == "__main__":
     unittest.main()

tests/test_sdk_peer_pinning.py

@@ -0,0 +1,294 @@
+from __future__ import annotations
+
+import socket
+import sys
+import threading
+import types
+import unittest
+from importlib.util import find_spec
+from http.server import BaseHTTPRequestHandler, ThreadingHTTPServer
+from unittest.mock import patch
+
+from govoplan_core.security.outbound_http import (
+    OutboundHttpBlocked,
+    create_outbound_connection as core_create_outbound_connection,
+)
+from govoplan_files.backend.storage.sdk_peer_pinning import (
+    SdkPeerPinningError,
+    _botocore_transport_types,
+    create_pinned_s3_client,
+    install_pinned_smb_transport,
+    pinned_smb_connection_cache,
+)
+
+
+class _Socket:
+    def __init__(self) -> None:
+        self.connected_to: object | None = None
+
+    def settimeout(self, _value: object) -> None:
+        return None
+
+    def setsockopt(self, *_args: object) -> None:
+        return None
+
+    def connect(self, value: object) -> None:
+        self.connected_to = value
+
+    def close(self) -> None:
+        return None
+
+
+class _S3Handler(BaseHTTPRequestHandler):
+    def do_GET(self) -> None:  # noqa: N802 - stdlib handler contract
+        body = (
+            b'<?xml version="1.0" encoding="UTF-8"?>'
+            b'<ListAllMyBucketsResult xmlns="http://s3.amazonaws.com/doc/2006-03-01/">'
+            b'<Owner><ID>govoplan</ID><DisplayName>GovOPlaN</DisplayName></Owner>'
+            b'<Buckets><Bucket><Name>evidence</Name><CreationDate>2026-08-04T00:00:00Z</CreationDate>'
+            b'</Bucket></Buckets></ListAllMyBucketsResult>'
+        )
+        self.send_response(200)
+        self.send_header("Content-Type", "application/xml")
+        self.send_header("Content-Length", str(len(body)))
+        self.end_headers()
+        self.wfile.write(body)
+
+    def log_message(self, _format: str, *_args: object) -> None:
+        return None
+
+
+@unittest.skipUnless(find_spec("botocore") and find_spec("boto3"), "boto3 extra is not installed")
+class S3PeerPinningTests(unittest.TestCase):
+    def test_client_is_born_with_the_pinned_http_session(self) -> None:
+        client = create_pinned_s3_client(
+            endpoint_url="http://127.0.0.1:9000",
+            region_name="eu-central-1",
+            aws_access_key_id="access",
+            aws_secret_access_key="secret",
+        )
+        try:
+            session_type, _, _ = _botocore_transport_types()
+            self.assertIsInstance(client._endpoint.http_session, session_type)
+            self.assertEqual({}, client._endpoint.http_session._proxy_config._proxies)
+        finally:
+            client.close()
+
+    def test_each_connection_attempt_resolves_and_pins_again(self) -> None:
+        _, http_connection, _ = _botocore_transport_types()
+        connection = http_connection(host="objects.example.test", port=80)
+        sockets = [_Socket(), _Socket()]
+        with patch(
+            "govoplan_files.backend.storage.sdk_peer_pinning.create_outbound_connection",
+            side_effect=sockets,
+        ) as connector:
+            self.assertIs(sockets[0], connection._new_conn())
+            self.assertIs(sockets[1], connection._new_conn())
+
+        self.assertEqual(2, connector.call_count)
+        self.assertEqual("objects.example.test", connector.call_args.args[0])
+
+    def test_real_sdk_request_uses_the_pinned_socket_path(self) -> None:
+        server = ThreadingHTTPServer(("127.0.0.1", 0), _S3Handler)
+        thread = threading.Thread(target=server.serve_forever, daemon=True)
+        thread.start()
+        endpoint = f"http://127.0.0.1:{server.server_port}"
+        try:
+            with patch.dict(
+                "os.environ",
+                {"APP_ENV": "production", "GOVOPLAN_CONNECTOR_ALLOW_PRIVATE_NETWORKS": "true"},
+            ), patch(
+                "govoplan_files.backend.storage.sdk_peer_pinning.create_outbound_connection",
+                wraps=core_create_outbound_connection,
+            ) as connector:
+                client = create_pinned_s3_client(
+                    endpoint_url=endpoint,
+                    region_name="eu-central-1",
+                    aws_access_key_id="access",
+                    aws_secret_access_key="secret",
+                )
+                try:
+                    response = client.list_buckets()
+                finally:
+                    client.close()
+        finally:
+            server.shutdown()
+            server.server_close()
+            thread.join(timeout=2)
+
+        self.assertEqual("evidence", response["Buckets"][0]["Name"])
+        self.assertEqual("127.0.0.1", connector.call_args.args[0])
+
+    def test_every_sdk_selected_origin_uses_a_pinned_pool(self) -> None:
+        session_type, http_connection, https_connection = _botocore_transport_types()
+        session = session_type(proxies={})
+        try:
+            first = session._manager.connection_from_url("http://one.example.test/root")
+            redirected = session._manager.connection_from_url("https://two.example.test/root")
+            self.assertIs(first.ConnectionCls, http_connection)
+            self.assertIs(redirected.ConnectionCls, https_connection)
+        finally:
+            session.close()
+
+    def test_tls_connection_retains_the_configured_authority_for_sni(self) -> None:
+        _, _, https_connection = _botocore_transport_types()
+        connection = https_connection(host="objects.example.test", port=443)
+        with patch(
+            "govoplan_files.backend.storage.sdk_peer_pinning.create_outbound_connection",
+            return_value=_Socket(),
+        ) as connector:
+            connection._new_conn()
+
+        self.assertEqual("objects.example.test", connection.host)
+        self.assertEqual("objects.example.test", connection._dns_host)
+        self.assertEqual("objects.example.test", connector.call_args.args[0])
+
+    def test_mixed_answer_and_peer_change_fail_closed(self) -> None:
+        _, http_connection, _ = _botocore_transport_types()
+        connection = http_connection(host="objects.example.test", port=80)
+        records = [
+            [(socket.AF_INET, socket.SOCK_STREAM, 6, "", ("93.184.216.34", 80))],
+            [
+                (socket.AF_INET, socket.SOCK_STREAM, 6, "", ("93.184.216.34", 80)),
+                (socket.AF_INET, socket.SOCK_STREAM, 6, "", ("10.0.0.5", 80)),
+            ],
+        ]
+        with patch.dict(
+            "os.environ",
+            {"APP_ENV": "production", "GOVOPLAN_CONNECTOR_ALLOW_PRIVATE_NETWORKS": "false"},
+        ), patch(
+            "govoplan_core.security.outbound_http.socket.getaddrinfo",
+            side_effect=records,
+        ), patch(
+            "govoplan_core.security.outbound_http.socket.socket",
+            return_value=_Socket(),
+        ):
+            connection._new_conn()
+            with self.assertRaisesRegex(Exception, "non-public network"):
+                connection._new_conn()
+
+
+class _FakeTcp:
+    def __init__(self, server: str, port: int, timeout: float | None = None) -> None:
+        self.server = server
+        self.port = port
+        self.timeout = timeout
+        self.connected = False
+        self._sock = None
+        self._sock_lock = threading.Lock()
+
+
+class _FakeConnection:
+    def __init__(self, _guid: object, server: str, port: int, *, require_signing: bool) -> None:
+        self.server = server
+        self.port = port
+        self.require_signing = require_signing
+        self.transport = _FakeTcp(server, port)
+        self.session_table: dict[str, object] = {}
+
+    def connect(self, *, timeout: float) -> None:
+        self.transport.timeout = timeout
+        self.transport.connect()
+
+
+class _FakeSession:
+    def __init__(
+        self,
+        connection: _FakeConnection,
+        *,
+        username: str | None,
+        password: str | None,
+        require_encryption: bool,
+        auth_protocol: str,
+    ) -> None:
+        self.connection = connection
+        self.username = username
+        self.password = password
+        self.encrypt_data = require_encryption
+        self.auth_protocol = auth_protocol
+        self.encrypt = require_encryption
+
+    def connect(self) -> None:
+        self.connection.session_table[self.username or "anonymous"] = self
+
+
+class SmbPeerPinningTests(unittest.TestCase):
+    def setUp(self) -> None:
+        pinned_smb_connection_cache.cache_clear()
+
+    def _modules(self) -> tuple[types.ModuleType, dict[str, types.ModuleType]]:
+        smbclient = types.ModuleType("smbclient")
+        pool = types.ModuleType("smbclient._pool")
+
+        def original_register_session(
+            server: str,
+            username: str | None = None,
+            password: str | None = None,
+            port: int = 445,
+            encrypt: bool | None = None,
+            connection_timeout: float = 60,
+            connection_cache: dict[str, object] | None = None,
+            auth_protocol: str = "negotiate",
+            require_signing: bool = True,
+        ) -> None:
+            del server, username, password, port, encrypt, connection_timeout
+            del connection_cache, auth_protocol, require_signing
+
+        pool.register_session = original_register_session
+        pool.ClientConfig = lambda: types.SimpleNamespace(client_guid="guid")
+        connection = types.ModuleType("smbprotocol.connection")
+        connection.Connection = _FakeConnection
+        connection.Tcp = _FakeTcp
+        session = types.ModuleType("smbprotocol.session")
+        session.Session = _FakeSession
+        transport = types.ModuleType("smbprotocol.transport")
+        transport.Tcp = _FakeTcp
+        return smbclient, {
+            "smbclient._pool": pool,
+            "smbprotocol.connection": connection,
+            "smbprotocol.session": session,
+            "smbprotocol.transport": transport,
+        }
+
+    def test_initial_reconnect_and_referral_hosts_are_each_pinned(self) -> None:
+        smbclient, modules = self._modules()
+        sockets = [_Socket(), _Socket(), _Socket()]
+        with patch.dict(sys.modules, modules), patch(
+            "govoplan_files.backend.storage.sdk_peer_pinning.create_outbound_connection",
+            side_effect=sockets,
+        ) as connector:
+            install_pinned_smb_transport(smbclient)
+            register = modules["smbclient._pool"].register_session
+            cache: dict[str, object] = {}
+            first = register("files.example.test", connection_cache=cache)
+            first.connection.transport.connected = False
+            register("files.example.test", connection_cache=cache)
+            register("dfs-target.example.test", connection_cache=cache)
+
+        self.assertEqual(
+            ["files.example.test", "files.example.test", "dfs-target.example.test"],
+            [call.args[0] for call in connector.call_args_list],
+        )
+
+    def test_referral_to_disallowed_peer_fails_before_session_creation(self) -> None:
+        smbclient, modules = self._modules()
+        with patch.dict(sys.modules, modules), patch(
+            "govoplan_files.backend.storage.sdk_peer_pinning.create_outbound_connection",
+            side_effect=OutboundHttpBlocked("non-public network"),
+        ):
+            install_pinned_smb_transport(smbclient)
+            register = modules["smbclient._pool"].register_session
+            with self.assertRaisesRegex(ValueError, "non-public network"):
+                register("private-referral.example.test", connection_cache={})
+
+    def test_transport_tampering_after_installation_fails_closed(self) -> None:
+        smbclient, modules = self._modules()
+        with patch.dict(sys.modules, modules):
+            install_pinned_smb_transport(smbclient)
+            modules["smbprotocol.connection"].Tcp = _FakeTcp
+            with self.assertRaisesRegex(SdkPeerPinningError, "changed after peer pinning"):
+                install_pinned_smb_transport(smbclient)
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_search_source.py

@@ -0,0 +1,145 @@
+from __future__ import annotations
+
+from types import SimpleNamespace
+import unittest
+
+from sqlalchemy import create_engine
+from sqlalchemy.orm import Session
+
+from govoplan_access.backend.db.models import Account, Group, User
+from govoplan_core.auth import ApiPrincipal
+from govoplan_core.core.access import PrincipalRef
+from govoplan_core.core.change_sequence import ChangeSequenceEntry
+from govoplan_core.core.events import EventObjectRef, EventTenantRef, PlatformEvent
+from govoplan_core.core.search import (
+    SearchAuthorizationRequest,
+    SearchBackfillRequest,
+    SearchResourceReference,
+)
+from govoplan_core.db.base import Base
+from govoplan_files.backend.db.models import FileAsset, FileFolder, FileShare
+from govoplan_files.backend.search_source import (
+    FilesSearchSource,
+    PROVIDER_ID,
+)
+
+
+class FilesSearchSourceTests(unittest.TestCase):
+    def setUp(self) -> None:
+        self.engine = create_engine("sqlite://")
+        Base.metadata.create_all(
+            self.engine,
+            tables=(
+                Account.__table__,
+                User.__table__,
+                Group.__table__,
+                FileAsset.__table__,
+                FileFolder.__table__,
+                FileShare.__table__,
+                ChangeSequenceEntry.__table__,
+            ),
+        )
+        self.session = Session(self.engine)
+        self.session.add_all(
+            (
+                Account(
+                    id="account-1",
+                    email="one@example.test",
+                    normalized_email="one@example.test",
+                ),
+                User(
+                    id="user-1",
+                    tenant_id="tenant-1",
+                    account_id="account-1",
+                    email="one@example.test",
+                ),
+                FileAsset(
+                    id="file-1",
+                    tenant_id="tenant-1",
+                    owner_type="user",
+                    owner_user_id="user-1",
+                    display_path="records/permit.pdf",
+                    filename="permit.pdf",
+                    description="Monthly permit evidence",
+                ),
+                FileAsset(
+                    id="file-other",
+                    tenant_id="tenant-2",
+                    owner_type="user",
+                    display_path="other.pdf",
+                    filename="other.pdf",
+                ),
+            )
+        )
+        self.session.commit()
+        self.source = FilesSearchSource()
+
+    def tearDown(self) -> None:
+        self.session.close()
+        self.engine.dispose()
+
+    def test_backfill_and_authorization_are_tenant_bounded(self) -> None:
+        page = self.source.backfill(
+            self.session,
+            request=SearchBackfillRequest(
+                tenant_id="tenant-1",
+                provider_id=PROVIDER_ID,
+                resource_type="file",
+                rebuild_id="rebuild-1",
+            ),
+        )
+        self.assertEqual(("file-1",), tuple(doc.resource_id for doc in page.documents))
+        reference = SearchResourceReference(
+            tenant_id="tenant-1",
+            module_id="files",
+            resource_type="file",
+            resource_id="file-1",
+        )
+        request = SearchAuthorizationRequest(reference=reference, source_revision="1")
+        self.assertTrue(
+            self.source.authorize(
+                self.session,
+                _principal({"files:file:read"}),
+                requests=(request,),
+            )[reference.key]
+        )
+        self.assertFalse(
+            self.source.authorize(
+                self.session,
+                _principal(set()),
+                requests=(request,),
+            )[reference.key]
+        )
+
+    def test_committed_file_event_produces_authoritative_upsert(self) -> None:
+        event = PlatformEvent(
+            type="files.file.updated",
+            module_id="files",
+            tenant=EventTenantRef(id="tenant-1"),
+            resource=EventObjectRef(type="file", id="file-1"),
+        )
+        changes = self.source.index_changes_for_event(
+            self.session,
+            event=event,
+            delivery_key="delivery-1",
+        )
+        self.assertEqual(1, len(changes))
+        self.assertEqual("upsert", changes[0].kind)
+        self.assertEqual(event.event_id, changes[0].cursor)
+
+
+def _principal(scopes: set[str]) -> ApiPrincipal:
+    return ApiPrincipal(
+        principal=PrincipalRef(
+            account_id="account-1",
+            membership_id="user-1",
+            tenant_id="tenant-1",
+            scopes=frozenset(scopes),
+        ),
+        account=SimpleNamespace(id="account-1"),
+        user=SimpleNamespace(id="user-1"),
+    )
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_share_lifecycle.py

@@ -0,0 +1,237 @@
+from __future__ import annotations
+
+import unittest
+from datetime import timedelta
+from unittest.mock import patch
+
+from sqlalchemy import create_engine
+from sqlalchemy.dialects import postgresql
+from sqlalchemy.orm import sessionmaker
+
+from govoplan_access.backend.db.models import Account, Group, User
+from govoplan_core.core.change_sequence import ChangeSequenceEntry
+from govoplan_core.db.base import Base
+from govoplan_files.backend.db.models import FileAsset, FileShare
+from govoplan_files.backend.storage.common import FileStorageError, utcnow
+from govoplan_files.backend.storage.files import (
+    _asset_visibility_query_for_user,
+    get_asset_for_user,
+    list_file_shares,
+    revoke_file_share,
+    share_file,
+)
+
+
+TENANT_ID = "tenant-1"
+OWNER_ID = "owner-1"
+RECIPIENT_ID = "recipient-1"
+GROUP_ID = "group-1"
+
+
+class FileShareLifecycleTests(unittest.TestCase):
+    def setUp(self) -> None:
+        self.engine = create_engine("sqlite:///:memory:", future=True)
+        Base.metadata.create_all(
+            bind=self.engine,
+            tables=[
+                Account.__table__,
+                User.__table__,
+                Group.__table__,
+                ChangeSequenceEntry.__table__,
+                FileAsset.__table__,
+                FileShare.__table__,
+            ],
+        )
+        self.session = sessionmaker(bind=self.engine, future=True)()
+        self.asset = FileAsset(
+            id="file-1",
+            tenant_id=TENANT_ID,
+            owner_type="user",
+            owner_user_id=OWNER_ID,
+            display_path="shared.pdf",
+            filename="shared.pdf",
+        )
+        self.session.add(self.asset)
+        self.session.commit()
+
+    def tearDown(self) -> None:
+        self.session.close()
+        self.engine.dispose()
+
+    @patch(
+        "govoplan_files.backend.storage.files.user_group_ids",
+        return_value=[],
+    )
+    def test_expired_share_stops_access_immediately(self, _groups) -> None:
+        self.session.add(
+            FileShare(
+                id="expired-share",
+                tenant_id=TENANT_ID,
+                file_asset_id=self.asset.id,
+                target_type="user",
+                target_id=RECIPIENT_ID,
+                permission="read",
+                expires_at=utcnow() - timedelta(seconds=1),
+            )
+        )
+        self.session.commit()
+
+        with self.assertRaisesRegex(FileStorageError, "No access"):
+            get_asset_for_user(
+                self.session,
+                tenant_id=TENANT_ID,
+                user_id=RECIPIENT_ID,
+                asset_id=self.asset.id,
+            )
+        self.assertEqual(
+            [],
+            list_file_shares(
+                self.session,
+                tenant_id=TENANT_ID,
+                asset_id=self.asset.id,
+            ),
+        )
+
+    @patch(
+        "govoplan_files.backend.storage.files.user_group_ids",
+        return_value=[GROUP_ID],
+    )
+    def test_independent_active_grant_survives_other_expiry(self, _groups) -> None:
+        self.session.add_all(
+            [
+                FileShare(
+                    id="expired-user-share",
+                    tenant_id=TENANT_ID,
+                    file_asset_id=self.asset.id,
+                    target_type="user",
+                    target_id=RECIPIENT_ID,
+                    permission="read",
+                    expires_at=utcnow() - timedelta(seconds=1),
+                ),
+                FileShare(
+                    id="active-group-share",
+                    tenant_id=TENANT_ID,
+                    file_asset_id=self.asset.id,
+                    target_type="group",
+                    target_id=GROUP_ID,
+                    permission="read",
+                    expires_at=utcnow() + timedelta(hours=1),
+                ),
+            ]
+        )
+        self.session.commit()
+
+        result = get_asset_for_user(
+            self.session,
+            tenant_id=TENANT_ID,
+            user_id=RECIPIENT_ID,
+            asset_id=self.asset.id,
+        )
+
+        self.assertEqual(self.asset.id, result.id)
+        self.assertEqual(
+            ["active-group-share"],
+            [
+                share.id
+                for share in list_file_shares(
+                    self.session,
+                    tenant_id=TENANT_ID,
+                    asset_id=self.asset.id,
+                )
+            ],
+        )
+
+    @patch(
+        "govoplan_files.backend.storage.files.ensure_share_target_exists",
+        return_value=None,
+    )
+    def test_revoke_is_idempotent_and_future_expiry_is_persisted(
+        self, _target_exists
+    ) -> None:
+        expiry = utcnow() + timedelta(days=1)
+        share = share_file(
+            self.session,
+            tenant_id=TENANT_ID,
+            asset=self.asset,
+            target_type="user",
+            target_id=RECIPIENT_ID,
+            permission="read",
+            user_id=OWNER_ID,
+            expires_at=expiry,
+        )
+        self.session.commit()
+
+        revoked, first_changed = revoke_file_share(
+            self.session,
+            tenant_id=TENANT_ID,
+            asset_id=self.asset.id,
+            share_id=share.id,
+            user_id=OWNER_ID,
+        )
+        self.session.commit()
+        repeated, second_changed = revoke_file_share(
+            self.session,
+            tenant_id=TENANT_ID,
+            asset_id=self.asset.id,
+            share_id=share.id,
+            user_id=OWNER_ID,
+        )
+
+        self.assertTrue(first_changed)
+        self.assertFalse(second_changed)
+        self.assertEqual(OWNER_ID, revoked.revoked_by_user_id)
+        self.assertEqual(revoked.revoked_at, repeated.revoked_at)
+        self.assertEqual([], list_file_shares(
+            self.session,
+            tenant_id=TENANT_ID,
+            asset_id=self.asset.id,
+        ))
+        self.assertEqual(
+            [share.id],
+            [
+                item.id
+                for item in list_file_shares(
+                    self.session,
+                    tenant_id=TENANT_ID,
+                    asset_id=self.asset.id,
+                    include_inactive=True,
+                )
+            ],
+        )
+
+    @patch(
+        "govoplan_files.backend.storage.files.ensure_share_target_exists",
+        return_value=None,
+    )
+    def test_past_expiry_is_rejected(self, _target_exists) -> None:
+        with self.assertRaisesRegex(FileStorageError, "future"):
+            share_file(
+                self.session,
+                tenant_id=TENANT_ID,
+                asset=self.asset,
+                target_type="user",
+                target_id=RECIPIENT_ID,
+                permission="read",
+                user_id=OWNER_ID,
+                expires_at=utcnow() - timedelta(seconds=1),
+            )
+
+    @patch(
+        "govoplan_files.backend.storage.files.user_group_ids",
+        return_value=[],
+    )
+    def test_visibility_query_is_postgresql_json_safe(self, _groups) -> None:
+        query = _asset_visibility_query_for_user(
+            self.session,
+            tenant_id=TENANT_ID,
+            user_id=RECIPIENT_ID,
+        )
+
+        compiled = str(query.statement.compile(dialect=postgresql.dialect()))
+
+        self.assertNotIn("SELECT DISTINCT", compiled.upper())
+        self.assertIn("EXISTS", compiled.upper())
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_storage_backends.py

@@ -1,19 +1,26 @@
 from __future__ import annotations
 
 import io
+from types import ModuleType
+import sys
 import unittest
 from unittest.mock import MagicMock, PropertyMock, patch
 
 from govoplan_files.backend.storage.backends import S3StorageBackend, StorageBackendError
 
 
-def _backend() -> S3StorageBackend:
+def _backend(
+    *,
+    endpoint_url: str = "https://objects.example.test",
+    deployment_managed: bool = False,
+) -> S3StorageBackend:
     return S3StorageBackend(
         bucket="files",
-        endpoint_url="https://objects.example.test",
+        endpoint_url=endpoint_url,
         region_name="test",
         access_key_id="access",
         secret_access_key="secret",
+        deployment_managed=deployment_managed,
     )
 
 
@@ -32,6 +39,48 @@ class S3StorageBackendTests(unittest.TestCase):
             ), self.assertRaisesRegex(StorageBackendError, "until that transport supports.*DNS/IP pinning"):
                 _backend().client
 
+    def test_installer_managed_garage_uses_exact_endpoint_and_path_style(self) -> None:
+        boto3 = ModuleType("boto3")
+        boto3.client = MagicMock(return_value=object())
+        botocore = ModuleType("botocore")
+        botocore.__path__ = []
+        botocore_config = ModuleType("botocore.config")
+
+        class Config:
+            def __init__(self, **values):
+                self.values = values
+
+        botocore_config.Config = Config
+        with patch.dict(
+            sys.modules,
+            {
+                "boto3": boto3,
+                "botocore": botocore,
+                "botocore.config": botocore_config,
+            },
+        ):
+            _backend(
+                endpoint_url="http://garage:3900",
+                deployment_managed=True,
+            ).client
+
+        _args, kwargs = boto3.client.call_args
+        self.assertEqual("http://garage:3900", kwargs["endpoint_url"])
+        self.assertEqual(
+            {"s3": {"addressing_style": "path"}},
+            kwargs["config"].values,
+        )
+
+    def test_installer_managed_garage_rejects_any_other_endpoint(self) -> None:
+        with self.assertRaisesRegex(
+            StorageBackendError,
+            "restricted to http://garage:3900",
+        ):
+            _backend(
+                endpoint_url="http://other-s3:3900",
+                deployment_managed=True,
+            ).client
+
     def test_get_bytes_rejects_declared_oversize_object_without_reading(self) -> None:
         body = MagicMock()
         client = MagicMock()