AgentSkillsCN

verification

在声称某项工作已完成、已修复,或已通过测试之前,可使用此技能——在做出任何成功声明前,需运行验证命令并确认输出结果。

SKILL.md
--- frontmatter
name: verification
description: Use when about to claim work is complete, fixed, or passing — requires running verification commands and confirming output before making any success claims

Verification Before Completion

The Iron Law

code
NO COMPLETION CLAIMS WITHOUT FRESH VERIFICATION EVIDENCE

If you haven't run the verification command in this message, you cannot claim it passes.

The Gate

code
BEFORE claiming any status:

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?
   - NO → State actual status with evidence
   - YES → State claim WITH evidence
5. ONLY THEN: Make the claim

Skip any step = lying, not verifying

What Requires Verification

ClaimRequiresNOT Sufficient
Tests passTest command output: 0 failuresPrevious run, "should pass"
Linter cleanLinter output: 0 errorsPartial check
Build succeedsBuild command: exit 0Linter passing
Bug fixedTest original symptom: passesCode changed, assumed fixed
Requirements metLine-by-line checklist verified"Tests passing"

Red Flags — STOP

If you catch yourself:

  • Using "should", "probably", "seems to"
  • Expressing satisfaction before verification ("Great!", "Done!")
  • About to commit/push without verification
  • Relying on a previous run
  • Thinking "just this once"

ALL of these mean: Run the verification command NOW.

Common Rationalizations

ExcuseReality
"Should work now"RUN the verification
"I'm confident"Confidence ≠ evidence
"Just this once"No exceptions
"Linter passed"Linter ≠ compiler ≠ tests
"Partial check is enough"Partial proves nothing

Key Patterns

Tests:

code
✅ [Run test command] → [See: 34/34 pass] → "All tests pass"
❌ "Should pass now" / "Looks correct"

Build:

code
✅ [Run build] → [See: exit 0] → "Build passes"
❌ "Linter passed so build should be fine"

Requirements:

code
✅ Re-read plan → Create checklist → Verify each item → Report gaps or completion
❌ "Tests pass, task complete"

The Bottom Line

Run the command. Read the output. THEN claim the result.

This is non-negotiable.