Applies Test-Driven Development principles to create verifiable, token-efficient, and highly discoverable Claude Code skills.
This skill treats documentation as an engineering discipline by applying the Red-Green-Refactor cycle to the creation of SKILL.md files. It ensures that instructions are not just written, but verified against baseline AI behaviors using pressure scenarios. By focusing on Claude Search Optimization (CSO) and token efficiency, this skill helps developers build a library of reusable patterns that Claude can easily find and execute without exhausting context limits.
주요 기능
01Standardized SKILL.md structure with YAML metadata guidance
02Pressure-test scenarios to verify AI compliance with new rules
031 GitHub stars
04Token-efficient writing patterns to save context window space
05TDD workflow for documentation (Red-Green-Refactor)
06Claude Search Optimization (CSO) for improved skill discovery
사용 사례
01Refactoring existing instructions to be more token-efficient and searchable
02Codifying a complex engineering pattern into a reusable Claude skill
03Validating that an AI agent follows specific constraints before deploying a skill