What problem does it solve? Writing production code before tests often leads to untested logic and regressions. This Skill enforces a disciplined test-driven development loop where failing tests are written first, minimal code makes them pass, and refactoring happens only while tests stay green. ## Core Features & Use Cases - RED-GREEN-REFACTOR Cycle: Structures work into three phases—write a failing test, make it pass with minimal code, then refactor without adding features. - Automatic Test Runner Detection: Identifies the correct test command from project markers such as pyproject.toml (pytest), package.json (npm test), go.mod (go test), Cargo.toml (cargo test), and build.gradle or pom.xml. - Naming Conventions: Provides should_X_when_Y style test naming examples for pytest, Jest/Vitest, and Go. - Use Case: When asked to implement a bulk status-change API, the Skill first writes a failing test asserting the expected behavior, verifies the failure is a genuine assertion failure, then implements the smallest passing change and refactors safely. ## Quick Start Ask the AI to build a feature test-first by saying: use TDD to implement the bulk status change API, starting with a failing test.
