Coordinate end-to-end Claude Code skill creation with evaluation loops and benchmark analysis.
npx skills add https://github.com/Aias/dotfiles --skill skill-creator-aias
Please help me install this Agent Skill. Skill: skill-creator Source: https://github.com/Aias/dotfiles/tree/main/.agents/skills/skill-creator Command: npx skills add https://github.com/Aias/dotfiles --skill skill-creator-aias
The Skill Creator guides users through drafting, testing, and refining Claude Code skills with an integrated feedback loop, enabling rapid skill iteration and quality assurance.
Provide a starting description for a new skill, then run the loop to iterate on triggers and performance.
High-intent search queries and answers about installing and using this skill.
To create and iterate on Claude skills, you draft a starting description, then use an automated evaluation loop to test prompts, benchmark performance, and refine triggers until the skill meets your automation workflow requirements.
An evaluation loop in prompt engineering automatically runs test prompts against your skill, analyzes benchmark results, and applies targeted improvements to the description and triggers, tracking history to measure iterative performance gains.
You structure a SKILL.md file with frontmatter and body instructions, then optionally include supporting resources like scripts, references, and assets to guide Claude Code behavior and define workflow automation triggers.
Yes, you use YAML for configuration and Anthropic dependencies to run automated test prompts, generate benchmark analysis reports, and identify strengths and weaknesses through post-hoc evaluation.
The best way to test skill triggers is to run a guided evaluation loop with holdout and test splits, review auto-generated benchmark reports, and refine the prompt engineering based on live feedback before deployment.
Skill prompts fail to trigger correctly when descriptions are vague or untested; running an automated evaluation loop generates benchmark reports that identify weaknesses and guide iterative improvements to triggers.
estebandrgCreate AI agent skills with SKILL.md frontmatter and template directories.
remcostoetenCritically evaluate plans and designs through structured iterative questioning.
HuuBarValidate Claude Code skills and generate improvement reports across review modes.
KILWA73Create Claude Skills with SKILL.md structure and evaluation loops.
