菜单
Use before claiming "done", "fixed", "tested", or "it runs", and before committing or opening a PR or MR—no task is exempt. Accept only evidence gathered after the last relevant change and just sufficient to prove the claim; do not over-verify small changes.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Use when creating a branch, using worktrees for parallel work, committing, or wrapping up. Covers branch naming, conventional commits, short-lived branches, worktree isolation, and final merge/cleanup.
Use before proposing a permanent fix for bugs, test failures, or unexpected behavior with an unknown root cause—mandatory when symptoms are far from the cause, a prior fix failed, or the issue crosses components. Reproduce, trace backward to the root cause, fix once at the source, and add a regression test. Do not trigger for obvious errors whose message identifies the cause (syntax/typo/missing import); fix those directly.
Use before coding when requirements have multiple reasonable interpretations, approaches require tradeoffs, or mistakes would be costly (permissions/payments/data migration/public APIs/data structures/hard-to-revert changes). Clarify requirements and design one question at a time, then get approval before implementation. Do not trigger for a focused small change with a clear goal and approach, or for mechanical edits (copy/constants/config values).
Use in cross-session work when user-confirmed outcomes, non-goals, hard constraints, or acceptance criteria have not been written back to the current source of truth (spec), or reality conflicts with it enough that following it would be wrong; also use during wrap-up or handoff when a source of truth should exist but does not. Edit only the affected original wording; do not keep separate old and new versions. Do not trigger for work that can be completed within one session, local implementation details, verification against existing acceptance criteria, locating the source of truth, or ordinary documentation edits.
Use before implementation when behavior can be verified by automated tests and regression coverage has clear value. For bugs with an unknown root cause, use systematic-debugging first, then return for red-green. Follow red-green-refactor—write a failing test first, add only enough code to pass it, then refactor. Do not trigger for documentation, configuration, copy-only, or styling changes with no testable behavior.
Use when changes affect interactions across multiple modules, involve high-risk areas (authorization/authentication, payments/funds, data deletion/migration, public APIs/cross-service interfaces/security boundaries), or the user requests a review. Review in this order—correctness → requirements → security → simplicity → style. Verify feedback before acting; do not comply blindly or agree performatively. Unless the user asks, documentation-only changes, explicit config values, and mechanical renames do not trigger merely because of a PR or merge.
| name | verify-before-done |
| description | Use before claiming "done", "fixed", "tested", or "it runs", and before committing or opening a PR or MR—no task is exempt. Accept only evidence gathered after the last relevant change and just sufficient to prove the claim; do not over-verify small changes. |
Hard rule: Only evidence gathered after the last relevant change to the code, files, or environment, and directly sufficient to prove the claim, counts.
Skipping any step = lying, not verification.
If a command cannot run (missing environment, broken dependencies, timeout), verification failed. Report that verification could not be completed and what blocked it. Do not quietly mark the step green as “skipped,” and do not misreport an environment problem as a product failure.
| Claimed evidence | Why it does not count |
|---|---|
| lint / type checks passed | This proves only that the checks pass, not that the project builds, runs, or behaves correctly. |
| A subagent reported "success" | That is its claim. Inspect the diff / actual artifact yourself. |
| The regression test passed once | This does not prove it catches the problem. In a safe, isolated way, temporarily disable this round's fix → the test must fail → restore the fix → the test must pass. Do not do this by reverting or overwriting user files. |
| "Tests are green = requirements are met" | Check each requirement explicitly; green tests alone are not enough. |
| Only part of the suite ran | This does not mean the whole suite passed. |
