skills.rest
run6270
run6270COMMUNITY
Explore all skills by this author →
GitHub

skill-creator

Create, modify, evaluate, and optimize AI skills with SKILL.md metadata and benchmarks.

1|Updated Feb 3, 2026
Search Tags:#skill development#prompt engineering#benchmarking#evaluation#ai agent#iterative refinement
One-click install
npx skills add https://github.com/run6270/skill --skill skill-creator-run6270
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/run6270/skill/tree/main/skill-creator
Command: npx skills add https://github.com/run6270/skill --skill skill-creator-run6270

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill streamlines the entire lifecycle of creating and refining AI capabilities, from initial concept to performance measurement, making AI development accessible and efficient.

Core Features & Use Cases

  • Skill Creation: Guide users through defining intent, writing SKILL.md, and structuring skill resources.
  • Iterative Improvement: Facilitate testing, evaluation, and refinement based on user feedback and performance metrics.
  • Performance Benchmarking: Run automated tests to compare skill versions and identify areas for optimization.
  • Description Optimization: Enhance skill triggering accuracy through targeted prompt engineering and evaluation.
  • Use Case: A user wants to build a new skill to summarize meeting transcripts. This Skill will help them draft the SKILL.md, set up test cases, run evaluations against baseline performance, analyze the results, and iterate on the skill's instructions and description until it performs optimally.

Quick Start

Use the skill-creator to help me build a new skill that can summarize meeting transcripts.

Frequently Asked Questions about skill-creator

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

FAQPage Schema
How do I build and test an AI agent skill from scratch?▼

To build an AI skill, define its intent, draft the SKILL.md metadata, bundle scripts and references, then run automated benchmarks to evaluate and iteratively refine its performance.

What is the best way to improve AI skill triggering accuracy?▼

Improve AI skill triggering accuracy through description optimization, using targeted prompt engineering and structured evaluation to refine how the skill responds to user requests.

Can I run automated benchmarks to compare different versions of my AI skills?▼

Yes, you can run automated benchmarks to compare skill versions, identify areas for optimization, and measure performance against baseline metrics to ensure iterative improvement.

How does iterative refinement work when developing prompt engineering skills?▼

Iterative refinement uses structured testing and feedback loops to evaluate prompt instructions, analyze test results, and apply modifications until the AI skill performs optimally.

Do I need prior coding experience to create and modify AI skills?▼

No extensive coding experience is required to start, as the skill guides users through defining intent, writing metadata, and structuring resources, making AI development accessible.

Related Skills

Explore More Software Engineering Skills→
s977043s977043
2

skill-creator

Design a Claude Code skill from a use case and produce a repo-ready package.

Community
Advanced
HeyvhuangHeyvhuang
339

skill-creator

Scaffold SKILL.md files with frontmatter and resource directories.

Community
Intermediate
AndyAnh174AndyAnh174

llm-evaluation

Evaluates LLM-powered systems using automated metrics, human feedback, and benchmarking.

Community
Advanced
mymx2mymx2
4

skill-creator

Draft, test, and benchmark AI skill instructions with automated evaluation loops.

Community
Advanced