agentleFS
quruhao4800quruhao4800/agent-workflow/skills/debugging/verification-before-completion/SKILL.md

verification-before-completion

Skill1 starsChanged 6 months agoMITView on GitHub

Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always

189 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use chinese-code-review, chinese-commit-conventions and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • Leodorareluctant259Skill

    chinese-code-review

    中文代码审查规范——在保持专业严谨的同时,用符合国内团队文化的方式给出有效反馈

    Also in 8 projects that use this

  • Leodorareluctant259Skill

    chinese-commit-conventions

    中文 Git 提交规范 — 适配国内团队的 commit message 规范和 changelog 自动化

    Also in 8 projects that use this

  • obraSkill

    diagnosing-superpowers

    Use when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.

    Also in 20 projects that use this

  • jh941213Skill

    docs-ci

    문서 CI(링크·OpenAPI lint·breaking 게이트·mermaid 검증)나 문서 드리프트 감지(docs.yaml 신선도 검사)를 설치하거나 돌릴 때 사용. 문서 내용 작성은 docs-architecture·docs-interfaces·docs-manuals.

    Also in 3 projects that use this

  • jh941213Skill

    docs-interfaces

    API·이벤트 인터페이스 문서(OpenAPI·swagger·AsyncAPI 명세, 엔드포인트 레퍼런스, API 구성도·시퀀스, API CHANGELOG)를 만들거나 코드와 맞출 때 사용. 시스템 구성도는 docs-architecture, API 설계 원칙은 api-design-principles.

    Also in 3 projects that use this

  • +3 more from Leodorareluctant259
  • +4 more from jh941213

The file

Show its rules and text

We split this skill into the rules it gives an agent

14 rules. 0 are common, 11 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 177 edited versions we read keep these.

  • 11

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Before any success/completion statement:

    line 25
  2. Distinctive:

    Run full command(s) now.

    line 28
  3. Distinctive:

    Read full output and exit code.

    line 29
  4. Distinctive:

    If known mistakes repeated without prevention checks, stop and fix process first.

    line 43
  5. Distinctive:

    Before completion claim, verify from plan artifacts:

    line 52
  6. Distinctive:

    Run coverage report and confirm thresholds are met:

    line 63
  7. Distinctive:

    If coverage is below threshold, add missing tests before claiming completion.

    line 74
  8. Very specific:

    No open blocking mistakes in 99-mistake-log.md

    line 82
  9. Very specific:

    No open CR-### or DR-### in 05-change-log.md

    line 83
  10. Distinctive:

    Before completing the Minimum Checklist, confirm the evidence matches the task type:

    line 87
  11. Very specific:

    Test run stdout: all tests passed (BUILD SUCCESSFUL, Tests run: N, Failures: 0, Errors: 0)

    line 90
  12. Distinctive:

    If the task adds or modifies an API endpoint: include an actual HTTP request + response (status code + body)

    line 94
  13. Distinctive:

    If multi-environment: list which environments the change has been applied to

    line 104
  14. Distinctive:

    Run commands. Read outputs.

    line 135

“Common” is measured: kept by at least 6 in 10 of the 177 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Verification Before Completion

Prerequisites

From skillMust already existIf missing
executing-plansAll tasks in 03-implementation-plan.md marked completedReturn to executing-plans — verification cannot start with open tasks
executing-plans04-verification.md updated with evidence for each completed taskReturn to executing-plans to fill evidence before running gates here

Overview

Claims require evidence. No completion claim is valid without fresh verification outputs.

The Iron Law

NO COMPLETION CLAIMS WITHOUT FRESH VERIFICATION EVIDENCE

Command Evidence Gate

Before any success/completion statement:

  1. Identify command(s) that prove the claim.
  2. Run full command(s) now.
  3. Read full output and exit code.
  4. State result with evidence only.

Mistake Recurrence Gate

Review:

  • docs/plans/YYYY-MM-DD-<feature-name>/99-mistake-log.md
  • docs/memory/global-mistakes.md

Research Recheck Gate

If 06-research-notes.md exists:

If known mistakes repeated without prevention checks, stop and fix process first.

  1. Recheck all entries marked Recheck Needed.
  2. Record recheck result in 04-verification.md.

Completion claims are invalid while required rechecks are pending.

Requirement Closure Gate (MANDATORY)

Before completion claim, verify from plan artifacts:

  • unmapped_req = 0 (all active REQ-### mapped to plan tasks and verification items)
  • open_cr = 0 (all CR-### closed or explicitly deferred with approval)
  • open_dr = 0 (all DR-### closed or explicitly deferred with approval)
  • pending_recheck = 0 (research rechecks done)

Record these values in 04-verification.md.

Coverage Gate

Run coverage report and confirm thresholds are met:

LayerMinimum
Service (business logic)85%
Controller80%
Overall project80%

Commands:

  • Java: ./gradlew test jacocoTestReport → check build/reports/jacoco/

If coverage is below threshold, add missing tests before claiming completion.

Minimum Checklist

  • Tests/build/lint evidence attached
  • Coverage thresholds met (evidence from report)
  • Requirement checklist verified line-by-line
  • 04-verification.md updated
  • No open blocking mistakes in 99-mistake-log.md
  • No open CR-### or DR-### in 05-change-log.md

Evidence Templates by Task Type

Before completing the Minimum Checklist, confirm the evidence matches the task type:

code-task

  • Test run stdout: all tests passed (BUILD SUCCESSFUL, Tests run: N, Failures: 0, Errors: 0)
  • JaCoCo coverage summary: Service ≥ 85%, Controller ≥ 80%, Overall ≥ 80%
    • Command: ./gradlew test jacocoTestReport
    • Report: build/reports/jacoco/test/html/index.html
  • If the task adds or modifies an API endpoint: include an actual HTTP request + response (status code + body)

migration-task

  • Flyway run log: Successfully applied N migration(s) to schema
  • Schema state after migration: output of DESCRIBE <table> or SHOW CREATE TABLE <table> showing expected columns and indexes
  • Rollback test: run rollback migration and confirm schema returns to pre-migration state (show output)

config-task

  • Application startup log showing configuration loaded without errors
  • Functional smoke test output confirming the affected behavior works
  • If multi-environment: list which environments the change has been applied to

api-endpoint (supplement to code-task)

Provide all three:

# Success path
Request:  POST /api/xxx
Body:     {"field": "value"}
Response: HTTP 200 {"code": 0, "data": {...}}

# Error path
Request:  POST /api/xxx
Body:     {"field": ""}
Response: HTTP 400 {"code": "INVALID_PARAM", "message": "..."}

# Auth rejection (if endpoint is protected)
Request:  POST /api/xxx (no token)
Response: HTTP 401

"It should work" or "tests cover this" is not evidence. Paste actual output.

Red Flags

  • "Should pass now"
  • "Looks good" without running commands
  • Declaring done while requirement mapping is incomplete
  • Delegated agent said success but no independent verification

Bottom Line

Run commands. Read outputs. Verify requirement closure. Then claim completion.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.