/test

Run tests with coverage reporting and quality gates.

Usage

/test [target] [--coverage] [--watch] [--failed] [--verbose]

Arguments

• target: Specific test file, directory, or test name (optional)
• --coverage: Generate coverage report (default: true)
• --watch: Watch mode for continuous testing
• --failed: Re-run only failed tests
• --verbose: Verbose output

Instructions

When this skill is invoked:

Agent Behavior (Codex-Max Pattern)

Autonomy:

• Complete test execution and analysis end-to-end
• If tests fail, analyze failures and suggest specific fixes
• If coverage is low, identify missing test cases and offer to implement them
• Don't just report results - provide actionable next steps

Thoroughness:

• Run full quality suite, not just tests
• Check prd/00_technology.md for project-specific commands:

  {lint_command}                    # Linting
  {type_check_command}              # Type checking
  {test_coverage_command}           # Tests with coverage

• Report on all aspects: tests, coverage, linting, type checking

Problem-Solving:

• If tests fail:
  1. Show the failure details with file:line references
  2. Analyze the root cause
  3. Suggest or implement the fix
  4. Re-run tests to verify
• If coverage is low:
  1. Identify uncovered lines (show file:line references)
  2. Suggest test cases to add
  3. Offer to implement the missing tests

Presentation:

• Lead with overall status (pass/fail, coverage %)
• Group results logically (by test file or category)
• Use file:line references for failures and missing coverage
• Provide clear next steps

Default Run (All Tests)

1. Read prd/00_technology.md to find the correct test commands

2. Run tests with coverage (example patterns):

   # See prd/00_technology.md for your specific command
   {test_runner} tests/ --coverage

3. Report results:

   ==================== Test Results ====================
   
   ✅ Passed: 42
   ❌ Failed: 0
   ⏭️  Skipped: 2
   ⏱️  Duration: 3.45s
   
   ==================== Coverage ====================
   
   Name                              Stmts   Miss  Cover
   -----------------------------------------------------
   src/{project}/__init__              2      0   100%
   src/{project}/main                 45      0   100%
   ...
   -----------------------------------------------------
   TOTAL                             523      0   100%
   
   ✅ Coverage requirement met

Specific Target

# Test a specific file
/test tests/unit/test_config

# Test a specific class/describe
/test tests/unit/test_api::TestHealthCheck

# Test a specific function/it
/test tests/unit/test_api::test_health_check

# Test a directory
/test tests/integration/

Coverage Analysis

1. If coverage < minimum:

   ⚠️  Coverage below requirement: 95% (required: {min_coverage}%)
   
   Missing coverage in:
   - src/{project}/services/user:45-52 (error handling)
   - src/{project}/api/auth:78-85 (edge case)
   
   Suggested tests to add:
   1. Test error handling in UserService.create()
   2. Test authentication timeout scenario

2. Generate detailed report if available in your stack

Integration Tests

When running integration tests:

1. Check dependencies are running (Docker services, etc.)
2. Run with integration marker from your test framework
3. Verify cleanup after tests

Quality Gates

The test skill enforces these quality gates (see prd/00_technology.md for specific values):

Gate	Requirement	Action on Failure
Coverage	Minimum %	Block merge
All tests pass	0 failures	Block merge
No skipped (without reason)	Documented skips only	Warn
Performance	No test > configured limit	Warn

Example Output

$ /test

🧪 Running tests...

tests/unit/test_config::TestSettings::test_default_values PASSED
tests/unit/test_config::TestSettings::test_is_development PASSED
tests/unit/test_api::test_health_check PASSED
tests/unit/test_api::test_root_endpoint PASSED

==================== 4 passed in 0.45s ====================

📊 Coverage: 100% ✅