$tdd

$tdd enforces the rule that a failing test must exist before any production code is written. It runs the red-green-refactor cycle stage by stage, consults the test-engineer agent for complex test strategies, and only exits a cycle when the new test passes and the entire existing test suite is green.

When to use it

• When you want to write tests first and attach implementation after
• When working on critical domain logic where correctness matters
• When you need systematic coverage instead of adding tests after implementation
• When learning a new codebase and want to document expected behavior with tests

Trigger keywords: "tdd", "test first", "red green", "write tests before code"

How to invoke

codex
> tdd add a coupon validation function to the checkout module

codex
> $tdd implement the password-strength checker with edge cases covered

How it works

RED — Write tests for the next unit of functionality and run them. If they pass on the first run, the test is considered wrong and must be rewritten.

GREEN — Write only the minimum production code needed to make the tests pass. The entire test suite must be green before moving on.

REFACTOR — Improve code quality while keeping tests green. Run tests after every change.

For complex domain logic or unfamiliar test patterns, the test-engineer agent reviews strategy before tests are written. Repeat the cycle until the feature is complete.

Outputs

• Failing tests written in the RED phase
• Minimal production code written in the GREEN phase
• Refactored source with the entire test suite green
• Per-cycle log of RED failure → GREEN pass → REFACTOR result

Related skills

• $autopilot — full execution pipeline that applies TDD discipline internally
• $ultraqa — QA cycle for fixing a broken test suite after implementation
• $ralplan — consensus planning that includes an expanded test plan in deliberate mode