关于
This skill establishes a rigorous framework for authoring Claude Code skills by treating documentation as code within a Test-Driven Development (TDD) cycle. It guides users through the process of writing pressure scenarios for subagents, implementing documentation to address observed failures, and refactoring for token efficiency and search optimization. By enforcing the mandate of having a failing test before writing any skill, it ensures that every new capability is verified against actual agent behavior, resulting in reliable, concise, and highly discoverable documentation for the Claude Code ecosystem.