What problem does it solve? Writing tests after implementation often produces brittle tests coupled to internal structure that break during refactors and fail to verify real behavior. This Skill enforces a disciplined red-green TDD loop so tests verify behavior through public interfaces and survive code changes. ## Core Features & Use Cases - Red-Green Loop Rules: Enforces writing a failing test first, then only the minimal code to pass it, one vertical slice at a time. - Seam-Based Test Design: Defines tests only at pre-agreed public interface seams, confirmed with the user before any test is written. - Anti-Pattern Detection: Identifies implementation-coupled, tautological, and horizontally-sliced tests with concrete good/bad examples. - Mocking Guidelines: Restricts mocks to system boundaries like external APIs, time, and file systems, with dependency injection patterns. - Use Case: When building a checkout feature, use this Skill to agree on the public seams, write a failing test like "user can checkout with valid cart", implement the minimal code to pass, and repeat slice by slice. ## Quick Start Ask the AI to build the next feature test-first using the tdd skill, starting by agreeing on the seams to test.