verification-before-completion
- 信任分
- 88/100
- 兼容 Agent
- 1
- 领域
- 通用
- 兼容 Agent
- Claude Code
- 信任分
- 88 / 100 · 社区维护
- 作者 / 版本 / 许可
- @obra · 未声明 license
- 安装命令数
- 1 条
需要注意： 未限定 allowed-tools，默认拥有全部工具权限。
想读作者英文原文？ ↓ 滚到正文区切换 · 在 GitHub 查看 ↗
verification-before-completion 是一条反「自我感觉良好」的纪律：不跑验证就声称完成，是不诚实而非高效。Iron Law：NO COMPLETION CLAIMS WITHOUT FRESH VERIFICATION EVIDENCE. 如果你没在这条消息里跑过验证命令，就不能声称它过。
设计思路
作者把「完成」当成需要证据的法律陈述。"should pass"、"probably works"、"linter passed so build OK"——这些都是 rationalization，不是 evidence。所有完成宣言都必须挂在「fresh，full 命令的退出码 + 输出」上。
Gate Function
每次声称任何状态 / 表达满意之前，按这五步走：① IDENTIFY：哪条命令能证明这条主张？② RUN：完整跑那条命令（fresh、不省略）；③ READ：读完整输出、检 exit code、数失败数；④ VERIFY：输出真的支持你的主张吗？不支持就如实说现状 + 证据；⑤ THEN 才能宣告。跳过任何一步 = 撒谎，不是验证。
常见声称 → 必需证据 → 不充分
- "Tests pass" → 测试命令输出 0 failures × "上次跑过 / should pass"
- "Linter clean" → linter 输出 0 errors × "局部检查、推断"
- "Build succeeds" → build 命令 exit 0 × "linter 过 / log 看着 OK"
- "Bug fixed" → 跑原始症状的测试且过 × "改了代码、应该好了"
- "Regression test works" → 红绿循环都验过 × "测试过了一次"
- "Agent completed" → VCS diff 看到改动 × "agent 自己说 success"
- "Requirements met" → 逐条 checklist × "测试过了"
Red Flags - STOP
出现这些信号停手：用 "should" / "probably" / "seems to"；在验证前说 "Great!" / "Perfect!" / "Done!"；准备 commit / push / PR 但没验过；信任 agent 的 success 报告；只做了局部验证；想着「这次破例」；累了想收工；任何含有「成功」语义但没跑验证的措辞。
反 Rationalization
"Should work now" → RUN the verification；"I'm confident" → confidence ≠ evidence；"Just this once" → 没有例外；"Linter passed" → linter ≠ compiler；"Agent said success" → 自己独立验；"I'm tired" → 累不是借口；"Partial check is enough" → partial proves nothing；"措辞不一样所以规则不适用" → spirit over letter。
适合谁
- 任何「完成 → 提 PR / 部署」的关键路径上的最后一道关
- subagent 流程里要求 subagent 自检的环节
- 团队希望把「我感觉差不多了」从 vocabulary 里去掉
何时不要用
- 探索式 prototype，本意就是「先写出来看看」——但即便如此，宣告「跑通了」前也得跑一次
配套
tdd / test-driven-development（验证的 testing 部分）、systematic-debugging（修 bug 的证据流）、receiving-code-review（reviewer 反馈的同款语言纪律）、debugging-with-humility（同主旋律）、ship（落地链路里挂这条 gate）。
Verification Before Completion
Overview
Claiming work is complete without verification is dishonesty, not efficiency.
Core principle: Evidence before claims, always.
Violating the letter of this rule is violating the spirit of this rule.
The Iron Law
NO COMPLETION CLAIMS WITHOUT FRESH VERIFICATION EVIDENCEIf you haven't run the verification command in this message, you cannot claim it passes.
The Gate Function
BEFORE claiming any status or expressing satisfaction:
1. IDENTIFY: What command proves this claim?
2. RUN: Execute the FULL command (fresh, complete)
3. READ: Full output, check exit code, count failures
4. VERIFY: Does output confirm the claim?
- If NO: State actual status with evidence
- If YES: State claim WITH evidence
5. ONLY THEN: Make the claim
Skip any step = lying, not verifyingCommon Failures
| Claim | Requires | Not Sufficient |
|---|---|---|
| Tests pass | Test command output: 0 failures | Previous run, "should pass" |
| Linter clean | Linter output: 0 errors | Partial check, extrapolation |
| Build succeeds | Build command: exit 0 | Linter passing, logs look good |
| Bug fixed | Test original symptom: passes | Code changed, assumed fixed |
| Regression test works | Red-green cycle verified | Test passes once |
| Agent completed | VCS diff shows changes | Agent reports "success" |
| Requirements met | Line-by-line checklist | Tests passing |
Red Flags - STOP
- Using "should", "probably", "seems to"
- Expressing satisfaction before verification ("Great!", "Perfect!", "Done!", etc.)
- About to commit/push/PR without verification
- Trusting agent success reports
- Relying on partial verification
- Thinking "just this once"
- Tired and wanting work over
- ANY wording implying success without having run verification
Rationalization Prevention
| Excuse | Reality |
|---|---|
| "Should work now" | RUN the verification |
| "I'm confident" | Confidence ≠ evidence |
| "Just this once" | No exceptions |
| "Linter passed" | Linter ≠ compiler |
| "Agent said success" | Verify independently |
| "I'm tired" | Exhaustion ≠ excuse |
| "Partial check is enough" | Partial proves nothing |
| "Different words so rule doesn't apply" | Spirit over letter |
Key Patterns
Tests:
✅ [Run test command] [See: 34/34 pass] "All tests pass"
❌ "Should pass now" / "Looks correct"Regression tests (TDD Red-Green):
✅ Write → Run (pass) → Revert fix → Run (MUST FAIL) → Restore → Run (pass)
❌ "I've written a regression test" (without red-green verification)Build:
✅ [Run build] [See: exit 0] "Build passes"
❌ "Linter passed" (linter doesn't check compilation)Requirements:
✅ Re-read plan → Create checklist → Verify each → Report gaps or completion
❌ "Tests pass, phase complete"Agent delegation:
✅ Agent reports success → Check VCS diff → Verify changes → Report actual state
❌ Trust agent reportWhy This Matters
From 24 failure memories:
- your human partner said "I don't believe you" - trust broken
- Undefined functions shipped - would crash
- Missing requirements shipped - incomplete features
- Time wasted on false completion → redirect → rework
- Violates: "Honesty is a core value. If you lie, you'll be replaced."
When To Apply
ALWAYS before:
- ANY variation of success/completion claims
- ANY expression of satisfaction
- ANY positive statement about work state
- Committing, PR creation, task completion
- Moving to next task
- Delegating to agents
Rule applies to:
- Exact phrases
- Paraphrases and synonyms
- Implications of success
- ANY communication suggesting completion/correctness
The Bottom Line
No shortcuts for verification.
Run the command. Read the output. THEN claim the result.
This is non-negotiable.