AgentSkillsCN

Coverage

覆盖率

SKILL.md

Test Coverage Analysis

You help analyze and improve test coverage across the codebase.

When to Use

Analyze test coverage when:

  • •Completing feature implementation
  • •Before merging pull requests
  • •Setting coverage goals
  • •Investigating low coverage areas
  • •Adding new features

How to Use

Say things like:

  • •"Analyze test coverage for the auth module"
  • •"Show me the coverage report"
  • •"Improve coverage for the user service"
  • •"Check what's not covered in the API routes"

Tools Available

  • •Run test coverage commands directly (jest --coverage, pytest --cov)
  • •Use read to examine coverage reports
  • •Use grep to find untested files
  • •Use write/edit to add tests

Coverage Goals

  • •Statement Coverage: Percentage of statements executed
  • •Branch Coverage: Percentage of branches taken
  • •Function Coverage: Percentage of functions called
  • •Line Coverage: Percentage of lines executed

Target Levels

  • •Critical Code: 90%+ coverage
  • •Important Code: 80%+ coverage
  • •Standard Code: 70%+ coverage
  • •Helper/Utility: 60%+ coverage

How to Improve Coverage

  1. •Analyze: Identify low-coverage areas
  2. •Prioritize: Focus on critical paths
  3. •Add Tests: Write test cases for uncovered code
  4. •Edge Cases: Add tests for edge cases
  5. •Verify: Run coverage report to confirm improvements

Example Workflow

bash
# Run coverage
npm test -- --coverage

# Check coverage report
cat coverage/lcov-report/index.html

# Identify gaps
# Look for files with < 70% coverage

# Add tests for uncovered code
# Write new test cases

Best Practices

  • •Set minimum coverage thresholds (e.g., 80%)
  • •Test both happy paths and error paths
  • •Use mocking for external dependencies
  • •Test edge cases and boundary conditions
  • •Keep tests fast and reliable
  • •Review coverage regularly