test-driven-development

Write one minimal failing test for a single behavior, run it with pytest to confirm it fails for the expected reason, then write the simplest code to make it pass. Finally refactor while keeping all tests green, and repeat the cycle for each new behavior.

Red means writing a failing test that defines the desired behavior. Green means writing the minimal code to make that test pass, even with hardcoded values. Refactor means cleaning up duplication and names while keeping all tests passing.

Tests written after code pass immediately, which proves nothing about whether they test the right thing. Watching a test fail first confirms it actually exercises the missing behavior and catches the bug it targets.

The Skill allows exceptions only for throwaway prototypes, generated code, and configuration files, and only after asking the user explicitly. Exploratory coding is fine, but the exploration must be discarded and rewritten with TDD.

An immediately passing test means you are testing existing behavior, not new functionality. Fix the test so it targets the missing feature, verify it fails for the expected reason, then proceed with implementation.

No, that horizontal slicing produces brittle tests designed before the implementation reveals the real interface. Use vertical tracer bullets instead: complete one red-green cycle per behavior before starting the next test.