/test

Legend

Symbol	Meaning		Abbrev	Meaning
→	leads to		E2E	end-to-end
&	and/with		MCP	model context protocol
w/	with		DB	database

Execute immediately. Add --plan flag if user wants to see plan first.

Create or run comprehensive tests for code in $ARGUMENTS.

Thinking flags (optional):

• --think→multi-component test planning
• --think-hard→complex test arch & strategy
• --ultrathink→comprehensive test system design

MCP flags (optional):

• --pup→browser-based E2E testing w/ Puppeteer | --c7→testing framework docs
• --seq→complex test scenario planning | --no-mcp→native test runners only

Examples:

• /user:test --e2e --pup --think - E2E tests w/ Puppeteer & context analysis
• /user:test --e2e --pw --think - E2E tests w/ Playwright UI verification
• /user:test --coverage --c7 - Test w/ framework docs support
• /user:test --integration --seq --think-hard - Complex test scenarios w/ deep planning

Pre-test cleanup:

• Clean test output dirs (coverage/, test-results/) | Remove test artifacts from previous runs
• Reset test DB/state→clean baseline | Clear test cache & temp files

When --coverage flag is present:

• Generate comprehensive coverage reports
• Identify untested code paths
• Aim for >80% code coverage
• Focus on critical business logic

When --watch flag is present:

• Set up automatic test execution on file changes
• Provide fast feedback during development
• Only run affected tests for efficiency
• Continuous integration during development

When --interactive flag is present:

• Interactive test runner with real-time feedback
• Allow test selection and filtering
• Step-through debugging for failing tests
• Live code coverage visualization

When --e2e flag is present:

• Create end-to-end tests for user flows
• Test critical paths through the application
• Include both happy paths and error scenarios
• Use appropriate E2E testing frameworks
• If --pw flag is present or UI verification is required, use Playwright:
  • Navigate key pages & critical flows
  • Capture snapshots/screenshots at key states
  • Check console for errors/warnings
  • Exercise core interactions (forms, navigation, modals)
  • Pause for user input when a decision is needed (auth, destructive action)

Test types to implement:

• Unit tests: Test individual functions, logic, and utilities in isolation
• Integration tests: Test APIs, services, and database interactions
• E2E tests: Test complete user workflows and critical paths

Testing Best Practices

Follow testing standards:

• Use descriptive test names that explain what is being tested
• Follow AAA pattern: Arrange, Act, Assert
• Use appropriate mocking for external dependencies
• Create fixtures for consistent test data
• Ensure tests are independent and can run in any order

Research requirements for testing:

• Testing frameworks → C7 documentation lookup for framework-specific patterns
• Test patterns → Must verify with official testing guides and best practices
• E2E testing → WebSearch for accessibility and user experience standards
• Performance testing → Research benchmarking patterns for the specific stack
• Never implement test patterns without documentation backing
• All test implementations must cite sources: // Source: [testing guide reference]

Report Output:

• Coverage reports: .claudedocs/metrics/coverage-<timestamp>.html
• Test summaries: .claudedocs/summaries/test-results-<timestamp>.md
• Ensure directory exists: mkdir -p .claudedocs/metrics/ .claudedocs/summaries/
• Include report location in output: "📄 Coverage report saved to: [path]"

Deliverables: Comprehensive test suite, coverage reports, testing documentation, and test strategy guide.