completion-gate
Use when about to say that work is done, fixed, passing, working, deployed or ready, before committing, opening or merging a pull request, closing a task, moving to the next task, or reporting a subagent's result as finished.
NoobyGains · 已有静态检查记录
安装与使用
运行命令前先阅读来源与依赖说明。复制命令不会自动安装或执行。
npx skills add NoobyGains/godmode --skill completion-gate技能说明
name: completion-gate description: "Use when about to say that work is done, fixed, passing, working, deployed or ready, before committing, opening or merging a pull request, closing a task, moving to the next task, or reporting a subagent's result as finished."
Claude Code arms this one-shot stop check when the skill is invoked and removes it
after its first pass; other harnesses ignore the field.
https://code.claude.com/docs/en/hooks#hooks-in-skills-and-agents
hooks: Stop: - hooks: - type: prompt once: true timeout: 30 prompt: >- The final assistant message passes the GodMode completion gate. It passes when it makes no claim that work is done, fixed, passing, working, complete, deployed, ready or verified; or when every such claim sits next to fresh evidence in that same message (the command or check that was run and its actual result: test counts, exit code, build output, response status, a diff, a screenshot); or when it says plainly what was not verified. It also passes whenever stop_hook_active is true. It fails only when a claim of that kind stands without evidence; then the reason must name the unsupported claim and tell the assistant to run the check that proves it now and show the output, or to restate exactly what is unverified.
Completion Gate
A claim about the state of the work is a statement of fact, and a fact needs evidence. The evidence is output you produced and read in this message: a test run, a build, a request and its response, a screenshot. If you did not run it just now, you do not know it.
The Prime Directive
NO CLAIM OF DONE, FIXED OR PASSING WITHOUT FRESH VERIFICATION OUTPUT IN THIS MESSAGE
No exceptions. No workarounds. No shortcuts.
When to Use
Required, before any of these:
- Saying or implying that something works, passes, is fixed, is complete or is ready, in any wording.
- Committing, pushing, opening a pull request, merging, tagging, deploying.
- Marking a task complete, moving to the next task, or ending your turn with a summary.
- Passing on a subagent's or another model's "done" as your own statement.
- Expressing satisfaction about the result ("that did it", "all good").
Skip:
- Statements that are explicitly about what you have not verified ("I have not run the tests yet"). Those are honest and need no gate.
The Protocol
-
Name the claim. Write down exactly what you are about to assert: "the login bug is fixed", "all tests pass", "the deploy uses the new database".
-
Name the proof. Decide which command or observation would prove that claim, and what output would disprove it. Use the table below. A claim about behaviour needs a check of behaviour; a check that something ran is not a check that it did the right thing.
-
Run it now, in full. The full test suite, not the one file you touched. The real build, not the linter. The real request against the running service, not the unit test that mocks it. Earlier runs, other sessions and a subagent's report do not count.
-
Read the output. The exit code, the counts (passed, failed, skipped, warnings), the error lines and the end of the log. Skipped tests and new warnings are findings, not noise.
-
Compare output to claim.
- The output supports the claim: state the claim with the evidence next to it.
- The output contradicts it: state the actual state with the evidence, and go back to work.
- The output is partial or ambiguous: say exactly what is verified and what is not.
-
Fill in the evidence table for any non-trivial completion and include it in the message:
Claim Command or check Result (from this message) Tests pass npm test214 passed, 0 failed, 0 skipped, exit 0 Build succeeds npm run buildexit 0, no warnings Bug fixed reproduction from the report returns 200 with the expected body; the test fails with the fix reverted Requirements met checklist against the plan, task 4 6 of 6 items checked, see below -
Only now make the claim, in plain words, with no hedging and no inflation.
What Counts as Proof
| Claim | Proof | Not proof |
|---|---|---|
| Tests pass | The full suite run now, zero failures, skips explained | An earlier run; "they should pass"; one file passing |
| Lint or types clean | The tool's output now, zero errors and warnings | Your editor shows no red lines |
| Build succeeds | The build command now, exit 0 | Lint passing; the dev server starting |
| Bug fixed | The original symptom reproduced and now absent | The code changed in the place you suspected |
| Regression test works | Red with the fix reverted, green with it restored | One green run |
| Requirements met | Each requirement checked off against the written spec | The tests pass |
| Subagent finished | The diff it produced, read by you, plus your own run of the checks | Its report saying "done" |
| Deployed | The live endpoint or the deployed version string observed | The deploy command exited 0 |
Configuration Changes
A configuration change can "succeed" because some valid configuration exists, while yours is ignored. This is the silent fallback: the request returns 200, but from the old provider; the app starts, but with the default flag; the job runs, but against staging.
So verify the difference, not the absence of errors:
- What should be different after the change? Name the observable value (a model name in the response, a hostname in the log, a feature visible on the page, an identity in the auth response).
- Where can you observe it? A response body, a log line, a header, a version endpoint, a query result.
- Run the command that exposes it and read the value.
- Confirm the value is the new one. If you only saw a status code, you have not verified the change.
|
阅读完整来源 ↗