Automate test-driven documentation generation for SKILL.md files.
npx skills add https://github.com/ayourtch/kimichat --skill writing-skills-ayourtch
Please help me install this Agent Skill. Skill: writing-skills Source: https://github.com/ayourtch/kimichat/tree/main/skills/writing-skills Command: npx skills add https://github.com/ayourtch/kimichat --skill writing-skills-ayourtch
This Skill ensures skills are effective, robust, and discoverable by applying Test-Driven Development (TDD) principles to their creation and refinement, making them bulletproof against rationalization.
systematic-debugging skill, use this skill to first run a scenario where an agent fails to debug systematically, then write the skill to address those failures.I'm using the writing-skills skill to create a new data-validation-patterns skill.
High-intent search queries and answers about installing and using this skill.
Test-driven documentation uses RED-GREEN-REFACTOR cycles: first run a scenario where an agent fails without the skill, then write the skill to pass that test, then refactor to close edge cases. This ensures documentation solves real problems rather than theoretical ones.
Skill descriptions need YAML frontmatter (name and description), a description starting with 'Use when' to clarify scope, rich keyword coverage for discoverability, and token-efficient phrasing. These elements maximize Claude Search Optimization and help users find your skill.
Bulletproofing techniques include explicit negations stating what the skill does not do, rationalization tables mapping common workarounds to why they fail, and red flags that highlight when someone might skip the skill. These block users from justifying non-use.
Yes. The RED-GREEN-REFACTOR framework applies to any skill creation: define a baseline failure scenario in your domain, write the skill to address it, then refactor to handle edge cases. The process is domain-agnostic; only the test scenarios change.
Without TDD, skills often contain aspirational content that sounds good but fails in real use. TDD-driven skills are validated against actual agent failure modes, making them robust, discoverable, and resistant to being bypassed or rationalized.
tense-iDispatch parallel agents to investigate independent failures across files or subsystems.
pradyumnasagarGuide authors through creating and validating ORS SKILL.md files.
nice-and-preciseApply test-driven development cycles to verify process documentation with subagents.
sei-newbearExecute TDD cycles with interface-first design and peer review.