Guide test-driven development with red-green-refactor loops and vertical tracer bullets.
npx skills add https://github.com/docs-plus/docs.plus --skill tdd-docs-plus
Please help me install this Agent Skill. Skill: tdd Source: https://github.com/docs-plus/docs.plus/tree/main/.agents/skills/tdd Command: npx skills add https://github.com/docs-plus/docs.plus --skill tdd-docs-plus
This Skill eliminates the common pain points of writing brittle tests that break during code refactors, implementing features with insufficient test coverage, and following inefficient horizontal slicing workflows that produce tests for imagined rather than actual user-facing behavior.
Ask the AI to implement the new user profile editing feature using test-driven development, starting with a test for successful profile updates with valid user input.
High-intent search queries and answers about installing and using this skill.
Unit tests break during refactoring when they are coupled to implementation details rather than observable behavior. Test-driven development resolves this by enforcing deep module interface design and appropriate system-boundary mocking to ensure tests remain refactor-safe.
Implement features using test-driven development by following the red-green-refactor loop with vertical tracer bullet cycles. Write a single test for actual user-facing behavior, implement minimal code to pass it, then iterate on additional behaviors like edge cases.
Horizontal slicing in TDD creates tests for imagined internal layers rather than actual user-facing behavior, producing fragile suites. Vertical tracer bullets instead slice through all layers to verify real observable behavior end-to-end, reducing test maintenance overhead.
Use mocks at the system boundary during integration testing to isolate external dependencies and verify actual user-facing behavior. Appropriate system-boundary mocking prevents implementation-coupled tests that fail during refactors while maintaining reliable test coverage.
Write refactor-safe tests for user authentication by starting with a single test for successful login, implementing minimal code to pass, then iterating on behaviors like password reset. This test-first approach verifies real observable behavior without coupling to internal structure.
qdzshAutomates red-green-refactor cycles for vertical slices with failing tests first.
vanillacake369Refactor code incrementally while preserving externally observable behavior.
kynopticGenerate fail-first feature tests for user-facing behaviors across languages and frameworks.
hculapEnforce red-green-refactor cycles with failing tests before production code.
