The QA Engineer agent is a specialized AI agent designed to help with test strategy, test planning, test automation, and comprehensive quality assurance for modern software projects.
Specialized agent for comprehensive software quality assurance. Helps design test strategies, implement unit/E2E tests with Playwright/Vitest, set up TDD workflows, and integrate automated testing into CI/CD pipelines.
/plugin marketplace add anton-abyzov/specweave/plugin install sw-testing@specweaveThe QA Engineer agent is a specialized AI agent designed to help with test strategy, test planning, test automation, and comprehensive quality assurance for modern software projects.
Invoke this agent when you need:
Use the Task tool with subagent_type:
Task({
subagent_type: "specweave-testing:qa-engineer:qa-engineer",
prompt: "Design a comprehensive test strategy for a React e-commerce application with Vitest unit tests, Playwright E2E tests, and accessibility testing"
});
JavaScript/TypeScript:
Other Languages:
Specialized Testing:
Testing Pyramid:
Test-Driven Development (TDD):
Behavior-Driven Development (BDD):
Prompt:
"Design a test strategy for a Next.js SaaS application with:
- User authentication (OAuth, email/password)
- Real-time dashboard with WebSocket
- Payment processing (Stripe)
- Admin panel with RBAC
- Multi-tenancy support
Target: 80%+ code coverage, < 5 min test execution, zero flaky tests"
What the agent provides:
Prompt:
"Set up Vitest for a TypeScript React project with:
- Unit tests for custom hooks
- Component tests with Testing Library
- API mocking with MSW
- Coverage thresholds (80%+ lines, 75%+ branches)
- Watch mode for development
- CI integration"
What the agent provides:
Prompt:
"Set up Playwright E2E tests for an e-commerce checkout flow:
- Product search and filtering
- Add to cart
- Checkout with guest and authenticated users
- Payment processing (test mode)
- Order confirmation
Cross-browser testing (Chrome, Firefox, Safari)
Mobile emulation (iPhone, Android)
Visual regression testing
Parallel execution"
What the agent provides:
Prompt:
"Guide me through TDD for implementing a shopping cart feature:
- Add item to cart
- Remove item from cart
- Update quantity
- Calculate total with tax
- Apply discount code
Use Vitest, follow red-green-refactor strictly"
What the agent provides:
Prompt:
"Set up automated accessibility testing for a React application:
- WCAG AA compliance
- axe-core integration in Vitest
- Playwright accessibility assertions
- Keyboard navigation tests
- Screen reader compatibility
- Color contrast validation
Target: Lighthouse accessibility score 95+"
What the agent provides:
Prompt:
"Set up performance testing for a REST API:
- Load testing (1000 concurrent users)
- Stress testing (find breaking point)
- Soak testing (24 hour sustained load)
- Metrics: p50, p95, p99 response times
Use k6, integrate with CI/CD, track trends"
What the agent provides:
npm install -D @playwright/testThe agent expects a clear, specific prompt describing:
Good prompt:
"Design test strategy for Node.js REST API with:
- 50+ endpoints (CRUD operations)
- Authentication (JWT)
- Rate limiting
- PostgreSQL database
- Redis caching
Stack: TypeScript, Express, Prisma
Target: 85%+ coverage, < 2 min test execution
CI: GitHub Actions
Team: 5 developers, all familiar with Jest"
Poor prompt:
"Need tests for my API"
The agent provides:
Focus on what code does, not how it does it.
No shared state between tests.
Unit tests should run in seconds.
Self-documenting test names and structure.
Page Object Model, fixtures, utilities.
Test against production-like environment.
80%+ code coverage, 100% critical paths.
Fix or quarantine flaky tests immediately.
When planning a feature increment:
# 1. Plan increment
/sw:increment "Implement user authentication with OAuth"
# 2. Generate test strategy
Task({
subagent_type: "specweave-testing:qa-engineer:qa-engineer",
prompt: "Design test strategy for OAuth authentication with Google, GitHub, and email/password. Include unit tests for auth logic, integration tests for OAuth flow, and E2E tests for complete user journey."
});
# 3. Implement with TDD
/sw:do
After implementation, validate test coverage:
# Run quality assessment
/sw:qa 0123
# Agent checks:
# - Test coverage (lines, branches, functions)
# - Test quality (independent, fast, readable)
# - Critical path coverage
# - E2E test completeness
# - Accessibility test coverage
Solution:
Solution:
Solution:
Task({
subagent_type: "specweave-testing:qa-engineer:qa-engineer",
prompt: `Set up Pact contract testing between:
- Frontend (React SPA)
- Backend API (Node.js)
- Third-party payment API (Stripe)
Ensure API compatibility across teams`
});
Task({
subagent_type: "specweave-testing:qa-engineer:qa-engineer",
prompt: `Set up visual regression testing with Percy for:
- 20+ pages across 3 breakpoints (mobile, tablet, desktop)
- 5+ themes (light, dark, high-contrast)
- Component library (Storybook integration)
CI integration, baseline management`
});
For issues or questions:
Designs feature architectures by analyzing existing codebase patterns and conventions, then providing comprehensive implementation blueprints with specific files to create/modify, component designs, data flows, and build sequences