Automate drafting, testing, and refining Claude skills with evaluations.
npx skills add https://github.com/family3253/skill --skill skill-creator-family3253
Please help me install this Agent Skill. Skill: skill-creator Source: https://github.com/family3253/skill/tree/main/skills/skill-creator Command: npx skills add https://github.com/family3253/skill --skill skill-creator-family3253
Enables teams to systematically create, test, and iteratively improve Claude skills from draft to deployment, ensuring reliable triggering and stable behavior while reducing the time to a production-ready skill.
Draft a new skill, run a small evaluation, review results, and iterate until triggering is stable.
High-intent search queries and answers about installing and using this skill.
Automate Claude skill evaluation by drafting a skill, running trigger evaluations, collecting metrics, and iterating on descriptions to improve triggering accuracy across varied prompts until behavior is stable.
Testing prompt triggering accuracy involves running trigger evaluations across diverse domains, aggregating benchmark results, and refining descriptions iteratively to ensure reliable skill deployment.
Validate frontmatter constraints during packaging by enforcing compatibility checks and skill validation rules, ensuring safe deployment-ready iterations with optional references and scripts included.
Yes, the skill benchmarking and evaluation workflow requires pyyaml as a dependency to parse configurations, enforce frontmatter constraints, and validate packaging for deployment-ready iterations.
Yes, you can run end-to-end skill creation automatically by guiding the workflow from initial draft through evaluation, results aggregation, and iterative refinement to produce a deployment-ready package.
Unreliable triggering across domains often indicates descriptions need refinement; run benchmark evaluations, collect metrics, and iterate on descriptions to improve triggering accuracy across varied prompts.
zts212653Generate and validate SKILL.md frontmatter and structure for new skills.
mobster570Generate and validate autonomous agent configurations for Claude Code plugins.
Shockwave2kDraft SKILL.md files and run evaluation loops to measure trigger accuracy.
dsteven12Create, test, and improve Cowork skills through automated evals and iteration.
