What problem does it solve? Writing tests after code produces tautological, implementation-coupled tests that pass while production breaks. This Skill enforces a disciplined red-green-refactor loop so every behavior is specified by a failing test before any production code exists. ## Core Features & Use Cases - Red-Green-Refactor Loop: Write one failing test per behavior, implement the minimal code to pass, then refactor only while the bar is green. - Seam Selection: Identify test boundaries (pure functions, ports/interfaces, HTTP boundaries, CLI entries) and confirm them with the user before writing any test. - Vertical Slicing: Build thin end-to-end slices per acceptance criterion instead of horizontal layer-by-layer implementation. - Anti-Pattern Detection: Catch implementation-coupled tests, tautological tests, horizontal slicing, and tests that only verify the framework. - Use Case: When implementing a new checkout feature from a spec, use this Skill to write a failing test for the first acceptance criterion, ship the minimal slice, and repeat until all criteria pass. ## Quick Start Use the tdd skill to implement the next acceptance criterion from my spec test-first, starting by proposing test seams for my approval.
