菜单
Use when the user explicitly asks to audit, critic, or verify a generated solution or answer (e.g. /verifying-solutions, "audit this", "critic this") — not for routine completion claims or pre-answer design.
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 the user asks to commit, write a commit message, generate a conventional commit, or invokes /conventional-commit. Also use when staged changes need a commitlint-compliant message, when the user asks to commit and push (or passes --push), when unstaged or untracked files exist and staging scope must be confirmed, or when agent-attribution trailers must be avoided on commits. Handles git commit and git push flags when the user passes them.
Use when the user asks to create, open, or raise a GitHub pull request, submit changes for review. Also when staged, committed, or unpushed work needs a PR via gh CLI and required PR metadata is missing (base branch, draft vs ready, reviewers, labels, linked issues, test plan).
Use when the task is multi-constraint, ambiguous, architectural, or needs a design/algorithm choice before answering. Skip for trivial lookups and one-line edits.
Use when staging, formatting, or creating git commits for code changes in a Git repository.
Analyzes error logs, stack traces, and provides systematic debugging strategies and solutions
Drafts professional emails based on context, tone, and purpose with templates and suggestions
| name | verifying-solutions |
| description | Use when the user explicitly asks to audit, critic, or verify a generated solution or answer (e.g. /verifying-solutions, "audit this", "critic this") — not for routine completion claims or pre-answer design. |
Secondary content audit of an already-produced answer or code. Explicit invoke only.
Does not replace structured-reasoning (pre-answer) or verification-before-completion (command evidence).
PASS → Audit Report only. FAIL → Audit Report + full Verified Solution.
"Always rewrite" does not override PASS. Polish is not a Verified Solution.
Score PASS/FAIL vs original requirements:
| Vector | Fail if |
|---|---|
| Constraint Matching | Any explicit requirement unmet |
| Edge Cases | Null/empty/boundaries/errors mishandled when in scope |
| Logical Correctness | Wrong behavior, fallacy, or incomplete logic |
| Syntactic/Reference Correctness | Broken syntax, bad imports, dangling refs |
Optimization: bullet suggestions only unless perf was required. Never FAIL for "prettier."
Test/build claims: Do not PASS from author notes. Require verification-before-completion evidence, or FAIL the claim.
Always emit the table (not freeform sections only):
### Audit Report
| Vector | Result | Notes |
|--------|--------|-------|
| Constraint Matching | PASS/FAIL | … |
| Edge Cases | PASS/FAIL | … |
| Logical Correctness | PASS/FAIL | … |
| Syntactic/Reference Correctness | PASS/FAIL | … |
**Verdict:** PASS | FAIL
[If FAIL: numbered failure points]
[Optional Optimization: bullets only — never a full alternate implementation]
Any hard FAIL → also:
### Verified Solution
[Full corrected production-ready solution — not a patch-only note]
All PASS → stop. No ### Verified Solution. No polish rewrite. No rewritten body under Optimization.
| Excuse | Reality |
|---|---|
| "Always provide Verified Solution" | PASS → Audit Report only |
| "Just polish / types / comments" | Polish ≠ rewrite |
| "Rewrite under Optimization on PASS" | Bullets only; full body only on FAIL |
| "Close enough / happy path" | Unmet constraints = FAIL + full fix |
| "Don't rewrite — meeting now" | Still FAIL + Verified Solution |
| "Tests pass / don't run anything" | No evidence → do not confirm |
| "Rubber-stamp is pragmatic" | Approving broken code is the failure |
### Verified Solution after Verdict PASSViolating the letter is violating the spirit.