What problem does it solve? Teams often write tests after implementation, producing biased tests that confirm existing behavior instead of defining expected behavior. This Skill enforces strict test-driven development: tests are written first, verified to fail (RED), then verified to pass after implementation (GREEN), with vertical slicing so only one test is in flight per cycle. ## Core Features & Use Cases - Framework auto-detection: Identifies Jest, Vitest, pytest, Cargo, Go test, Cypress, or Playwright from config files and follows existing project conventions. - Vertical TDD enforcement: Tracks bulk_test_count, requires paired test/feat commits per cycle, and blocks horizontal slicing where multiple tests are written before the first GREEN. - 4-layer test methodology: Covers unit, integration, true-backend, and E2E/subprocess layers, plus coverage checks (80% minimum), spec-to-test traceability, and eval scenarios for skill behavior testing. - Use Case: When implementing a password reset feature, the Skill writes one failing test for the email-sending behavior, hands off implementation, verifies GREEN, commits the pair, then moves to the next behavior such as rate limiting. ## Quick Start Ask the agent to write TDD tests for the new password reset feature following the project's existing test framework.