skills.rest
YouMingYeh
YouMingYehCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

skill-creator

Create and iterate on Claude skills with structured documentation and evaluation.

2|Updated Jan 12, 2026
Search Tags:#automation#documentation#evaluation#iteration#benchmark#skill-creation#prompt-engineering
One-click install
npx skills add https://github.com/YouMingYeh/shadcn-ui-supabase-docker-template --skill skill-creator-youmingyeh
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/YouMingYeh/shadcn-ui-supabase-docker-template/tree/main/.agents/skills/skill-creator
Command: npx skills add https://github.com/YouMingYeh/shadcn-ui-supabase-docker-template --skill skill-creator-youmingyeh

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires anthropic, pyyaml, and includes assets (resource) components.

What problem does it solve?

This Skill helps teams rapidly create, iterate, and evaluate Claude skills by automating evaluation loops, tracking iterations, and surfacing actionable improvement prompts.

Core Features & Use Cases

  • End-to-end skill lifecycle: draft SKILL.md, run evaluations, analyze results, and iterate to improve triggering accuracy.
  • Automated benchmarking: generate and interpret benchmark data to compare skill variants with/without baseline.
  • Descriptor optimization: leverage evaluation results to craft better SKILL.md frontmatter and body for better triggering.

Quick Start

Compose an initial SKILL.md and an eval set, then run the evaluation loop to begin iterative improvement.

Frequently Asked Questions about skill-creator

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I create and iterate on Claude skills with automated evaluation?▼

You create and iterate on Claude skills by drafting SKILL.md content, defining evals, and running automated evaluation loops to analyze results and improve triggering accuracy.

What is the best way to benchmark prompt-engineering skill variants?▼

The best way to benchmark skill variants is to generate and interpret benchmark data with and without baselines, comparing results to surface actionable improvement prompts for refinement.

How does automated skill evaluation work for descriptor optimization?▼

Automated skill evaluation works by running evaluation loops against defined eval sets, tracking iteration results, and leveraging those results to craft better SKILL.md frontmatter and body content.

Do I need anthropic and pyyaml dependencies to run skill evaluations?▼

Yes, you need the anthropic and pyyaml dependencies to run the skill evaluation loops, as they provide the necessary SDK components and configuration parsing for the automation lifecycle.

Why does my Claude skill fail to trigger accurately in production?▼

Your Claude skill may fail to trigger accurately because the SKILL.md frontmatter and body descriptors lack optimization; running automated benchmarking and iterating based on eval results improves triggering.

Related Skills

Explore More Software Engineering Skills→
coreyjacoreyja
12

skill-writer

Draft SKILL.md frontmatter and structure guidance for Claude Code skills.

Community
Basic
Citrus-bitCitrus-bit
135

skill-creator

Orchestrates skill creation, evaluation, benchmarking, and description optimization workflows for Claude Skills.

Community
Advanced
j1nggj1ngg
55

autoresearch

Benchmark and improve existing repo skills through repeated evals and score tracking.

Community
Advanced
BaryoDevBaryoDev

skill-creator

Create and iteratively refine AI skills with SKILL.md and performance evaluations.

Official
Advanced