tdd-dev-workflow

Create a structured GitHub issue (epic, bug, feature, discussion, question)

Central TDD orchestrator — drives the full development lifecycle from issue analysis through verified implementation

Browse open issues, claim one, and create a development branch

TDD-aligned comprehensive code review against GitHub issue acceptance criteria

Create a pull request with auto-generated title, linked issue, and configured reviewers

Use this agent for TDD-aligned code review against GitHub issue acceptance criteria. Can dispatch the debugger agent when bugs are found. <example> Context: TDD cycle and e2e tests complete for issue #42. Time for code review before PR. user: "Implementation is done. Review the code." assistant: "Dispatching code-reviewer agent to review against issue #42 acceptance criteria." <commentary>Code review validates implementation completeness and quality against the issue's requirements.</commentary> </example> <example> Context: User wants a standalone code review of their current branch. user: "Review my changes on this branch" assistant: "Dispatching code-reviewer agent to review all changes against origin/main." <commentary>The code-reviewer works both as part of the orchestrator and as a standalone review tool.</commentary> </example>

Use this agent for systematic debugging when a bug is found during code review or testing. Follows the 4-phase debugging methodology. <example> Context: The code-reviewer found a critical bug where transaction amounts are calculated incorrectly for multi-currency projects. user: "The reviewer found a calculation bug in transactions" assistant: "Dispatching debugger agent to systematically investigate the transaction calculation bug." <commentary>The debugger agent applies the 4-phase systematic debugging methodology to find and fix the root cause.</commentary> </example> <example> Context: E2E tests fail because the approval workflow doesn't trigger for high-value transactions. user: "E2E test failure: approval not triggered" assistant: "Dispatching debugger agent to investigate why the approval workflow fails for high-value transactions." <commentary>Test failures trigger systematic debugging — not random fix attempts.</commentary> </example>

Use this agent to write and run end-to-end tests that validate complete user workflows. Dispatched after TDD cycle completes. <example> Context: Unit tests pass for the new invoice creation feature. E2E tests need to verify the full user flow. user: "Unit tests pass. Now write e2e tests for the invoice creation flow." assistant: "Dispatching e2e-tester agent to write Playwright tests covering the complete invoice creation workflow." <commentary>E2E tests validate the full user journey through the actual UI after unit/integration tests pass.</commentary> </example> <example> Context: A bug fix changed the login redirect behavior. E2E tests should verify the full authentication flow still works. user: "The login fix is done. Verify the auth flow end-to-end." assistant: "Dispatching e2e-tester agent to write E2E tests validating the complete login and redirect flow." <commentary>After bug fixes that affect user-facing behavior, E2E tests confirm the fix works through the full stack.</commentary> </example>

Use this agent to implement minimal code that makes failing tests pass. Dispatched after tdd-red agent completes. <example> Context: The tdd-red agent has written 12 failing tests for a transaction approval feature. user: "Tests are written and failing. Now implement." assistant: "Dispatching tdd-green agent to write the minimal implementation to pass all 12 tests." <commentary>The green phase implements just enough code to make tests pass — no more.</commentary> </example> <example> Context: A bug was found during code review. The debugger identified the root cause and a new failing test was written to reproduce it. user: "Bug reproduction test is failing. Fix the implementation." assistant: "Dispatching tdd-green agent to implement the minimal fix that makes the bug reproduction test pass." <commentary>Bug fixes also go through the green phase — write just enough to make the reproduction test pass.</commentary> </example>

Use this agent to write failing tests for the RED phase of TDD. Dispatched by the develop orchestrator with acceptance criteria. <example> Context: The orchestrator has analyzed a GitHub issue and extracted acceptance criteria for a user authentication feature. user: "Implement issue #42 - Add user authentication" assistant: "Starting TDD cycle. Dispatching tdd-red agent to write failing tests for all acceptance criteria." <commentary>The tdd-red agent is dispatched as the first step in the TDD cycle to define expected behavior through tests.</commentary> </example> <example> Context: After a bugfix review loop, new tests are needed for the discovered edge case. user: "The reviewer found a bug in input validation" assistant: "Dispatching tdd-red agent to write a failing test that reproduces the validation bug." <commentary>Bug reproduction starts with a failing test — the red phase of the fix cycle.</commentary> </example>

This skill should be used when the user asks about "e2e testing", "end-to-end tests", "when to write e2e tests", "e2e test strategy", "what should I e2e test", "e2e test structure", "browser testing strategy", "data-testid selectors", "flaky e2e tests", "e2e vs unit tests", "Cypress", or needs guidance on when, what, and how to structure end-to-end tests. For Playwright-specific API patterns and configuration, see playwright-patterns instead. Provides E2E testing strategy and principles.

This skill should be used when the user asks about "issue lifecycle", "PR workflow", "project board", "development process", "merge strategy", "issue status", "kanban flow", "start working on issue", "pick up an issue", "create a PR", "submit a pull request", "resume work", "continue where I left off", or needs guidance on the GitHub issue-to-merge workflow. Orchestrates the full issue-to-merge lifecycle.

This skill should be used when the user asks about "NestJS", "Nest.js", "NestJS modules", "dependency injection", "guards", "interceptors", "pipes", "middleware", "NestJS testing", "TypeORM", "Prisma with NestJS", "NestJS authentication", "JWT guards", "NestJS validation", or needs guidance on NestJS architecture and patterns.

This skill should be used when the user asks about "Next.js", "App Router", "Server Components", "Client Components", "route handlers", "server actions", "streaming", "parallel routes", "intercepting routes", "ISR", "SSG", "SSR", "Next.js middleware", "Next.js caching", "Next.js metadata", or needs guidance on Next.js 14/15+ architecture and patterns.

This skill should be used when the user asks about "Supabase Auth with Next.js", "authentication Next.js", "login Supabase", "auth middleware Next.js", "protected routes", "auth callback", "Supabase SSR auth", "session management Next.js", "@supabase/ssr", or needs to integrate Supabase authentication in a Next.js App Router application.