Review checklist for correctness, regressions, and missing tests in the vault-memory FastAPI/async codebase — async pitfalls, FastAPI/DI/auth checks, and test-quality signals.
67
81%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
View guide
Passed
No findings from the security scan
await is exception-safe where failure is expected.app.dependency_overrides over source-inspection tests.asyncio.Task handles must be
(a) exception-safe inside the coroutine (try/except + log, so the exception
is always retrieved), (b) cleared in finally so a failed run re-arms,
(c) loop-aware — compare task.get_loop() against asyncio.get_running_loop()
before trusting task.done(), because a not-yet-done task from a dead loop
is never done.await between check and assign.asyncio.create_task(...) results must be kept
referenced (assign to a variable/collection) or the task may be
garbage-collected mid-flight.asyncio.run() inside running loops; in sync tests prefer
TestClient or asyncio.run only at top level.Depends(get_dependencies) pattern; mock via
app.dependency_overrides[get_dependencies].verify_api_key reads VAULT_MEMORY_API_KEY; unset env = dev mode.
Tests must monkeypatch-delete the key to avoid cross-test leakage.detail (error_response handles this);
expected user errors should be 4xx with a stable code string._safe_vault_path; validate that relative_to/resolve guards exist where new
paths are constructed.inspect.getsource) are acceptable as cheap
regression guards but must not be the only coverage for a branch.skipif(os.name == "nt")._bulk_jobs, _cleanup_task) must save
and restore the original values to avoid cross-test pollution.8171052
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.