What problem does it solve? It prevents common AI coding failures such as writing tests that merely mirror the implementation, skipping the Red verification step, over-testing internal details, and letting hardcoded values or tangled responsibilities slip into production code. ## Core Features & Use Cases - TDD Cycle Enforcement: Mandates the Explore → Red → Green → Refactor loop, requiring the failing test to actually be executed and confirmed before implementation begins. - Test Discipline Rules: Treats tests as specifications, bans implementation-fitted and tautological tests, limits test scope to agreed public seams, and requires documenting what the test suite does and does not guarantee. - Code Design Principles: Enforces separation of concerns, single responsibility, guard clauses, no hardcoding, security-first input handling, and delegating statically checkable rules to linters or ast-grep. - Use Case: When asking an AI agent to add a feature or fix a bug, this skill ensures it writes a failing test first, implements the minimum to pass, refactors cleanly, and documents the guarantee scope of the tests. ## Quick Start Apply the development-style rules while implementing this feature using test-driven development.
