Design, test, and iterate Claude Code skills from SKILL.md to evals.
npx skills add https://github.com/binfen1/my-skills --skill skill-creator-binfen1
Please help me install this Agent Skill. Skill: skill-creator Source: https://github.com/binfen1/my-skills/tree/main/claude-code/skill-creator Command: npx skills add https://github.com/binfen1/my-skills --skill skill-creator-binfen1
The Skill Creator provides a repeatable, end-to-end workflow to design, implement, evaluate, and iteratively improve Claude Code skills—from drafting SKILL.md entries to running evaluations, reviewing results, and applying improvements to triggering and descriptions.
Draft your SKILL.md, then run the evaluation loop to test, review, and iterate toward improvement.
High-intent search queries and answers about installing and using this skill.
To iterate and test Claude Code skills, this Skill provides an end-to-end workflow from drafting SKILL.md to running evaluations, reviewing results in a built-in UI, and applying improvements to triggers and descriptions.
The best way to optimize skill descriptions and reduce false positives is using the built-in description optimizer, which refines triggering accuracy based on evaluation results within the iteration loop.
You structure a SKILL.md file by capturing intent, defining trigger conditions, and creating a structured entry with name, description, compatibility, and references using the provided workflow.
Yes, the skill creation workflow supports running automated evals by allowing you to gather edge cases and success criteria during research, then benchmarking and reviewing results in the evaluation UI.
Yes, you need pyyaml installed as a dependency to use this Skill Creator, which leverages scripts and references to package and evaluate Claude Code skills for deployment.
etewe333Coordinate end-to-end Claude skill design from intent capture to SKILL.md drafting and eval-driven iteration.
rodmena-limitedCreate, configure, and manage Claude Code Skills with templates and tools.
Shockwave2kDraft SKILL.md files and run evaluation loops to measure trigger accuracy.
chenziyang110Create and iteratively improve Claude Code skills with trigger evaluations.
