# Design: saas-test-pilot **Domain:** `saas` **Target Version:** v9.5.0 ## Purpose Test automation toolkit supporting unit, integration, and end-to-end testing. Generates test cases from code analysis, manages test fixtures, and provides coverage analysis with gap detection. ## Target Users - Developers writing tests for Python or JavaScript/TypeScript projects - Teams enforcing test coverage requirements - Projects needing test generation from existing code ## Commands | Command | Description | |---------|-------------| | `/test setup` | Setup wizard — detect test framework, configure paths | | `/test generate` | Generate test cases for functions/classes/modules | | `/test coverage` | Analyze test coverage and identify untested code paths | | `/test fixtures` | Generate or manage test fixtures and mocks | | `/test e2e` | Generate end-to-end test scenarios from user stories | | `/test run` | Run tests with formatted output and failure analysis | ## Agent Architecture | Agent | Model | Mode | Role | |-------|-------|------|------| | `test-architect` | sonnet | acceptEdits | Test generation, fixture creation, e2e scenarios | | `coverage-analyst` | haiku | plan | Read-only coverage analysis and gap detection | ## Skills | Skill | Purpose | |-------|---------| | `framework-detection` | Detect pytest/Jest/Vitest/Playwright, identify config | | `test-patterns` | Unit/integration/e2e test patterns and best practices | | `mock-patterns` | Mocking strategies for different dependency types | | `coverage-analysis` | Coverage gap detection and prioritization | | `fixture-management` | Fixture organization, factories, builders | | `visual-header` | Standard command output headers | ## MCP Server **Not required.** Test generation is file-based. Test execution uses the project's own test runner via Bash. ## Integration Points | Plugin | Integration | |--------|-------------| | projman | `/sprint test` delegates to test-pilot when installed | | saas-api-platform | API route tests generated from `/api test-routes` | | saas-react-platform | Component tests generated alongside components | | data-seed | Test fixtures use seed data profiles | | code-sentinel | Security test patterns included in generation | ## Token Budget | Component | Estimated Tokens | |-----------|-----------------| | `claude-md-integration.md` | ~700 | | Dispatch file (`test.md`) | ~200 | | 6 commands (avg) | ~3,600 | | 2 agents | ~1,200 | | 6 skills | ~2,500 | | **Total** | **~8,200** | ## Open Questions - Should `/test run` replace projman's `/sprint test run` or supplement it? - Support for property-based testing (Hypothesis, fast-check)?