AgentSkillsCN

Testing

测试

SKILL.md
--- frontmatter
name: "Testing"
id: "testing"
version: "1.0"
claude_subagent: "test-runner"
fallback: "self-check"
min_tad_version: "2.1"
platforms: ["claude", "codex", "gemini"]

Testing Skill

Purpose

Execute and verify test suites, ensure adequate test coverage, and validate that all tests pass before code integration.

When to Use

  • •After implementing new features
  • •After fixing bugs
  • •Before Gate 3 verification
  • •During CI/CD pipelines

Checklist

Critical (P0) - Must Pass

  • • All existing tests pass without failures
  • • No test timeouts or hanging tests
  • • Test output is clean (no unexpected warnings)
  • • Critical paths have test coverage

Important (P1) - Should Pass

  • • New code has corresponding unit tests
  • • Test coverage meets project minimum (default: 70%)
  • • Integration tests pass for affected modules
  • • Edge cases are tested

Nice-to-have (P2) - Informational

  • • Tests are well-documented with clear descriptions
  • • Test data is properly isolated
  • • Performance-sensitive tests have benchmarks
  • • Tests follow project naming conventions

Pass Criteria

LevelRequirement
P0All items pass
P1Max 1 failure
P2Informational

Evidence Output

Path: .tad/evidence/reviews/{date}-testing-{task}.md

Execution Contract

  • •Input: file_paths[], test_command, coverage_threshold
  • •Output: {passed: bool, test_results: [], coverage: number, evidence_path: string}
  • •Timeout: 300s
  • •Parallelizable: false

Claude Enhancement

When running on Claude Code, call subagent test-runner for deeper analysis. Reference: .claude/skills/ for extended test guidance.

Platform-Specific Execution

Claude Code

code
Use Task tool with subagent_type: "test-runner"

Codex CLI / Gemini CLI

code
1. Read this SKILL.md checklist
2. Run project test command (npm test, pytest, etc.)
3. Verify each P0/P1 item manually
4. Generate evidence file with results