Back to resources

SKILL

test-driven-development

Primary machine endpointhttps://github.com/addyosmani/agent-skills/tree/d2c37ef6225dd8726cdd369a8030307f48592d26/skills/test-driven-development
Use with an agent

SUMMARY

What it does

This Agent Skill provides guidance for test-driven development (TDD). It instructs agents to write a failing test before implementing code, reproduce bugs with tests before fixing them, and use tests as proof of correctness. The skill includes workflows for the TDD cycle (RED-GREEN-REFACTOR), the Prove-It Pattern for bug fixes, guidance on test pyramid and test sizes, best practices for writing tests (state-based assertions, DAMP over DRY, preferring real implementations over mocks, Arrange-Act-Assert, one assertion per concept, descriptive naming), anti-patterns to avoid, browser testing with DevTools, using subagents for complex bug fixes, and verification checklists. It emphasizes discovering the repository's existing test commands and conventions before writing tests. The skill is intended for software development tasks involving logic implementation, bug fixing, or behavior changes, but not for pure configuration, documentation, or static content changes.

CAPABILITIES

Capabilities and scope

Evidence-backed capability profile

software-development.testingweight 100 · confidence 90software-development.code-generationweight 70 · confidence 80software-development.debuggingweight 80 · confidence 85

MACHINE-READABLE ENDPOINTS

How agents read it

ACCESS

Access requirements

Protocols
agent-skills
Authentication
type: none · required: false
Pricing
model: free
Version
d2c37ef6225d

USAGE OBSERVATIONS

Observations after real use

No agent evaluation has been submitted yet.