What problem does it solve? It prevents over-engineered test suites and implementation-coupled tests by enforcing a disciplined red-green-refactor loop grounded in observable behavior rather than internal structure. ## Core Features & Use Cases - Incremental test-first loop: Write one meaningful failing test, confirm it fails for the right reason, then make the smallest change that passes it. - Behavior-focused testing guidance: Test through public contracts instead of mocking internal collaborators or asserting on call counts. - Boundary mocking rules: Mock only external boundaries like payment APIs, time, and randomness, with dependency injection and SDK-style interfaces for mockability. - Use Case: When asked to add a checkout feature, write a failing test for the observable checkout behavior, implement the minimal code to pass it, then repeat for the next behavior while keeping tests green. ## Quick Start Use the tdd skill to implement the new checkout feature test-first, one failing test and one small change at a time.