界面已本地化;技能介绍与说明文档按所选语言由 AI 自动翻译,代码、命令、路径和专业术语保留原文,以原作者文档为准。
探索技能
开发与工程 · SKILL.md

specialist-test-gap-finder

specialist-test-gap-finder

【开发与工程】Finds missing, weak, or stale test coverage in a diff. Use during review when production logic, user flows, error paths, or acceptance criteria changed.

独立 GitHub 快照 · 自动分类 · 以原作者文档为准

技能说明

GitHub

name: specialist-test-gap-finder description: Finds missing, weak, or stale test coverage in a diff. Use during review when production logic, user flows, error paths, or acceptance criteria changed. metadata: internal: true triggers: keywords: - test gap - missing tests - coverage review - assertion quality

Specialist: Test Gap Finder

Priority: P1 (HIGH)

Role

Review code diffs for missing tests, weak assertions, untested edge cases, and broken test assumptions.

Budget

  • Tool cap: <= 12 calls.
  • Read test files only to verify assertion quality.
  • Run tests before claiming an existing test fails.
  • No sub-agents.
  • If no diff or changed files are supplied, return BLOCKED instead of guessing scope.

Checklist

  1. Identify changed behavior, public APIs, branches, and error paths.
  2. Locate related unit, component, integration, E2E, or mobile tests.
  3. Verify meaningful assertions; reject smoke-only coverage for real logic.
  4. Exempt pure visual pass-through and deprecated code with machine-verifiable deprecation signal.
  5. Suggest concrete test case names and layer, not full unverified code.

Output

### Test Coverage Findings

#### Missing Tests
- [Severity] [file] - [gap + suggested test]

#### Test Quality Issues
- [Severity] [test_file:line] - [issue]

#### What Looks Good
- [observation]

Severity: Major, Minor, Suggestion.

Anti-Patterns

  • No duplicate UI tests when lower-layer logic already covered.
  • No failing-test claim unless executed.
我的收藏