What problem does it solve? Code changes often ship without proof they work, bugs get fixed without failing tests that reproduce them, and recurring regressions resurface because prior failure context was never recovered. This Skill enforces a verification-first discipline so every fix and feature is backed by executable tests. ## Core Features & Use Cases - Red-Green-Refactor Workflow: Enforces writing a failing test before any implementation, then minimal code to pass, then safe refactoring under green tests. - Regression Memory Loop: Recovers prior failure evidence and decision trails before writing a red test, so recurring bugs are reproduced from evidence rather than guesswork. - Testing Tactics & Strategy: Covers the testing pyramid (unit, integration, E2E), mocking rules (stubs, spies, mocks), AAA pattern, and Playwright vs Cypress guidance. - AI Output Verification: Provides checklists for validating AI-generated code, including hallucination checks, confidence scoring, and mandatory build/lint/test gates. - Use Case: When asked to fix a production bug, the Skill directs you to first write a failing test that reproduces the exact failing input, then implement the minimal fix, then refactor with the test staying green. ## Quick Start Ask the assistant to fix a bug or add a feature using the TDD cycle, starting with a failing test that proves the behavior before writing any implementation code.
