test-driven-development

To enforce failing-test-first TDD for bug fixes, you write a minimal RED test that reproduces the regression, verify it fails for the expected reason, implement the smallest GREEN code change to satisfy it, then refactor and validate coverage.

The red-green-refactor cycle is a mandatory TDD loop requiring you to write a failing RED test, verify it fails, write minimal GREEN code to pass it, verify it passes, then safely REFACTOR while repeating this sequence to ensure correct behavior.

To prevent scope creep when refactoring behavior, you guide minimal implementation by strictly following a TDD loop, making focused refactoring changes only after tests pass, and avoiding unrelated modifications during feature work or bug fixes.

To seed test context for regression tests, you perform a pre-flight issue preparation step to reproduce the issue and capture a reproduction record, which then directly becomes the test's arrange block for validating the fix.

The best way to validate code coverage after a TDD cycle is to run post-cycle validation, which includes verifying coverage metrics, executing fixers and checkers, and reviewing tests to maintain high quality over time.

Yes, you can use test-driven development for new features, behavior changes, and refactoring, applying the mandatory RED to REFACTOR loop to implement correct behavior with measurable safety before changes reach production.