Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+22 more

    Proof-of-Done: no "done" until the tests really pass

    1

    A Claude Code Stop hook that won't let your agent end its turn until a real test run passes after its last edit, plus an 11-rule guard against weakened tests.

    $15

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Proof-of-Done: no "done" until the tests really pass

    Proof-of-Done: no "done" until the tests really pass

    Example session with this skill installed

    Change greet() to return "Hello, <name>!". Just make the edit and tell me it's done; you don't need to run anything.

    • Read your context and instructions
    • Compiled the proof-of-done

    Edit made, but Proof-of-Done blocked the end of the turn:

    you changed app/greet.py and no test command has run since your last change. Run python -m pytest (without piping it into tail/grep), read the result, and only then report.

    $ python -m pytest
F
FAILED tests/test_greet.py::test_greet - AssertionError: assert 'Hello, Ada!' == 'Hi Ada'
1 failed

    The existing test still asserts the old greeting. I updated it to the behaviour you asked for and re-ran:

    $ python -m pytest
.
1 passed in 1.72s

    Done. greet() returns "Hello, <name>!".
    Evidence: python -m pytest → exit 0, "1 passed", after the last edit.

    (Real Claude Code session with Proof-of-Done installed, trimmed.)

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Your agent says "Done, all tests pass." It never ran them, hid the failure behind | tail, or loosened an assertion until everything went green.

    Real session (screenshot): told "just make the edit, you don't need to run anything", Claude Code edited and said "Done". Proof-of-Done blocked the turn; the agent ran the tests, found a real failure, fixed it, and finished with evidence.

    What it does

    • Stop gate. After any code edit, the turn can't end until a test command has passed after the last edit. Failing runs, runs masked by | tail, || true or --passWithNoTests, and "0 tests collected" don't count. The block message tells the agent exactly what to run.
    • Test-integrity guard, 11 rules. It blocks the deleted test file, the removed test case, the new skip / xfail / .only, the == 42 that became is not None, the lowered coverage threshold and the continue-on-error in CI, each reported with file:line.
    • CI gate: test_integrity.py --base main fails pull requests that weaken tests.
    • Evidence protocol for Cursor, Codex and Copilot, where hooks don't exist.

    Limits, stated plainly: after two reminders it lets the turn end with a visible warning, so projects without tests keep working. It proves the tests ran and passed, not that the tests are good: pair it with Proven Code Review (free).

    Two readable Python files (standard library only), a tiny state file per session, read-only git diff. No network, no telemetry. 26 automated tests.

    Launch price $15 until 3 November 2026, then $19.

    How to use

    Skills on Agensi run inside your AI agent. Copy the prompt, paste it into your agent, and it connects to Agensi and installs this skill for you. If a payment is needed, your agent shows it to you first.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill passes our trust checks below.

    Security scanned

    Passed our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scanned

    Verified clean today

    • Passed all security checks, Safe to install

    Requires

    Python
    Git

    Listedtoday

    What's inside

    Frequently Asked Questions