Add can_access_table check to GET /api/catalog/profile/{table_name} and
POST /api/catalog/profile/{table_name}/refresh, returning 403 for
unauthorized tables. Update test_api_complete to cover new 403 behaviour
and fix the existing 404 test to use admin token.
Add require_role(Role.ADMIN) to /admin/tables and /admin/permissions,
and require_role(Role.KM_ADMIN) to /corporate-memory/admin so that
non-admin users receive 403 instead of being served the page.
Fix admin_cookie test fixture to supply a password_hash (required since
the /auth/token endpoint blocks passwordless requests). Add analyst
fixture and TestAdminRoleGuards tests verifying analysts get 403 and
admins get 200 on the protected routes.
Users without a password_hash (Google OAuth / magic-link accounts) could
obtain a JWT by simply posting their email to /auth/token. Add an else
clause that rejects such requests with 401, directing them to their
configured auth provider. Update and extend tests accordingly.
Add information_schema, duckdb_* introspection functions, pragma_* functions,
and relative path traversal patterns to the SQL blocklist so users cannot
enumerate schema metadata regardless of RBAC. Add six corresponding tests.
Removed kbcstorage from all dependency groups (optional + dev) so urllib3
is no longer pinned to <2.0. Legacy Keboola client is available via
manual install: pip install kbcstorage
kbcstorage pins urllib3<2.0.0 which blocks Dependabot security patches.
Moved to [project.optional-dependencies] keboola-legacy since the primary
extraction path uses the DuckDB Keboola extension, not kbcstorage.
Legacy fallback uses lazy import — app works without it installed.
- Removed deploy-production job — Kamal config has placeholder IPs, no secrets
- Renamed workflow to "Build & Push" — test + Docker image to GHCR
- Added FORCE_JAVASCRIPT_ACTIONS_TO_NODE24=true to suppress Node.js 20 warnings
- All dependencies now in pyproject.toml [project.dependencies]
- Dev/test deps in [project.optional-dependencies] dev and [tool.uv]
- Dockerfile uses uv pip install . from pyproject.toml
- CI uses uv pip install ".[dev]"
- Deleted requirements.txt and requirements-dev.txt
- Updated README, CLAUDE.md install instructions
- Enhanced .dockerignore (exclude tests, docs, infra from image)
- architecture.md rewritten for v2 (FastAPI, DuckDB, Docker) — removed
all Flask/rsync/SSH/systemd references
- Deleted PLAN.md and REFACTORING_PLAN.md (completed, superseded)
- auto-install.md replaced with redirect to DEPLOYMENT.md
- Fixed absolute paths in superpowers plan doc
- tests/test_web_ui.py: smoke tests for all authenticated web pages (login, dashboard, catalog, corporate-memory, activity-center, admin/tables, admin/permissions)
- tests/test_jira_service.py: unit tests for extract_init and update_meta in the Jira connector
- tests/test_instance_config.py: verifies get_instance_name() returns a string when config file is absent
- tests/test_orchestrator.py: concurrent rebuild test asserting rebuild succeeds while a read-only connection holds the analytics DB
The Flask-based webhook endpoint at connectors/jira/webhook.py is no longer used.
FastAPI handles webhooks via app/api/jira_webhooks.py which is already integrated
into the application. This removes the redundant Flask code.
Prevents environment variable leaking between tests. All DATA_DIR,
JWT_SECRET_KEY, and SCRIPT_TIMEOUT assignments in fixtures now use
monkeypatch.setenv() which auto-reverts after each test. Removes
manual os.environ.pop() cleanup lines.
Replace copy-pasted _get_data_dir() functions in catalog.py and upload.py
with import from app.utils.get_data_dir(). sync.py and data.py already use
the shared utility.
- Add close_system_db() function in src/db.py to cleanly close shared DB connection
- Add lifespan context manager in app/main.py to trigger shutdown on app exit
- Integrate lifespan into FastAPI app initialization
- All API tests pass (77/77)
- Add --timeout=60 to pytest.ini to catch hanging tests
- Add --strict-markers to enforce declared markers only
- Install pytest-timeout>=2.0.0 dependency
- All 627 tests pass within 60s timeout
Replace substring matching with word-boundary regex in query endpoint's
table access validation. Prevents false positives where short table names
like 'id' would block any query containing the word. Uses re.escape() to
safely handle special characters in table names.
- Import re module at top
- Use regex pattern with word boundaries (\b) for matching
- Add tests to verify no false positives and proper blocking
Write to extract.duckdb.tmp, then atomically swap into place with WAL cleanup.
Prevents lock conflicts with orchestrator holding read lock on existing database.
- In src/orchestrator.py _attach_remote_extensions: escape token with '' before passing to ATTACH
- In connectors/keboola/extractor.py _try_attach_extension: escape token with '' before passing to ATTACH
Prevents SQL injection if token contains single quotes.
DuckDB has used WAL by default since v0.8, so this pragma is not
valid DuckDB syntax. Removed obsolete try-except block that attempted
to enable WAL on system database initialization.
Replace inherited env vars with a minimal env dict (PATH, DATA_DIR, HOME only),
omitting VIRTUAL_ENV and PYTHONPATH to prevent subprocess access to installed
packages. Switch subprocess invocation to sys.executable so the correct
interpreter is used with the restricted PATH. Add httpx to blocked_patterns
and BLOCKED_MODULES. Add test_sandbox_cannot_import_httpx to test_security.py.
Adds _SAFE_IDENTIFIER regex guard before ATTACHing extract.duckdb files in the
read-only analytics connection, matching the same fix already applied in the
orchestrator. Adds test coverage for malicious directory names.
Add _atomic_swap_db helper that removes stale WAL files before and after
moving the temp DuckDB into place. Apply CHECKPOINT before close in both
orchestrator and Keboola extractor so DuckDB flushes WAL before the swap.
Tokens previously lasted 30 days with no revocation path. Expiry is now
24 hours and every token carries a unique jti (UUID hex) to support future
revocation checks.
Expand blocked keywords to cover parquet_scan, read_csv_auto, query_table,
iceberg_scan, delta_scan, call, URL schemes (http/https/s3/gcs), and
additional file-scan functions. Set enable_external_access=false on the
non-read-only analytics connection path. Add three new tests covering
parquet_scan, read_csv_auto, and query_table blocking.
Prevents production deployments from silently using a hardcoded default
secret. TESTING=1 still resolves to a built-in test key so the existing
test suite is unaffected. Adds a test that verifies the RuntimeError is
raised when neither JWT_SECRET_KEY nor TESTING is set.
Adds _validate_identifier() with ^[a-zA-Z_][a-zA-Z0-9_]{0,63}$ regex and
applies it to source_name (directory names), table_name (_meta rows), and
alias/extension (_remote_attach rows) before any SQL interpolation.
Adds two tests covering SQL-injection directory names and malicious _meta entries.
_do_rebuild_source was creating a fresh temp DB with only one source,
then atomically replacing analytics.duckdb — wiping views from every
other source. Now it delegates to _do_rebuild so all extract dirs are
re-attached in a single pass.
Adds test_rebuild_source_preserves_other_sources to guard the regression.
Previously the password check was gated on both user.password_hash and
request.password being truthy, so an attacker could omit the password
field (which defaults to "") and receive a valid JWT. Now any user with a
stored hash must supply a non-empty password that passes argon2 verification.
Adds six TestTokenEndpoint tests covering empty, missing, wrong, and correct
password, plus no-hash user and unknown user cases.
- Replace padak/tmp_oss → keboola/agnes-the-ai-analyst in all docs, infra, CLI
- Replace your-org/ai-data-analyst → keboola/agnes-the-ai-analyst in README, Jira docs
- Remove real GCP project ID from terraform.tfvars.example
- Delete internal draft documents (dev_docs/draft/)
- Update infra/main.tf to clone from main branch
- Legacy extractor now uses read_csv(all_varchar=true) to avoid type
inference errors (e.g. seniority column typed as DOUBLE with string values)
- DEPLOYMENT.md rewritten based on actual dev VM deployment experience:
deploy key setup, DuckDB write locking, env reload gotchas, bootstrap flow
deploy-guard.yml referenced deleted tests and sudoers files.
deploy.yml.example used legacy SSH-based deployment.
Updated ci.yml and deploy.yml are in .gitignore (need workflow scope to push).
