skills.rest
thechrickenrun-cyber
thechrickenrun-cyberCOMMUNITY
Explore all skills by this author →
GitHub

skill-creator

Create, evaluate, and package Claude skills from SKILL.md workflows.

1|Updated May 5, 2026
Search Tags:#skill creation#benchmarking#claude skills#packaging#iteration#evals
One-click install
npx skills add https://github.com/thechrickenrun-cyber/wedding --skill skill-creator-thechrickenrun-cyber
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/thechrickenrun-cyber/wedding/tree/main/.agents/skills/skill-creator
Command: npx skills add https://github.com/thechrickenrun-cyber/wedding --skill skill-creator-thechrickenrun-cyber

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Skill Creator helps you turn an idea into a working Claude skill, then improve it with real tests instead of guesswork.

Core Features & Use Cases

  • Skill drafting: Write or revise SKILL.md frontmatter and body content so the skill triggers for the right user intent.
  • Test design and evaluation: Create realistic eval prompts, run comparison tests, and grade outputs so you can see where the skill succeeds or fails.
  • Iteration and analysis: Review benchmark results, spot flaky or non-discriminating assertions, and refine the skill based on evidence.
  • Packaging and delivery: Validate the skill folder and package it into a distributable .skill file for sharing or installation.

Quick Start

Use this skill to turn a rough idea into a draft SKILL.md, a few realistic eval prompts, and a packaged skill file.

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 package a Claude skill from a rough idea?▼

You can build a Claude skill by drafting a SKILL.md file with frontmatter, generating realistic eval prompts, running benchmark loops, and packaging the validated folder into a .skill file.

What is the best way to evaluate and iterate on Claude skills?▼

The best way to evaluate Claude skills is to generate realistic eval prompts, run comparison benchmark tests, grade outputs, and refine the skill based on analyzed evidence and results.

Do I need pyyaml to build and test Claude skills?▼

Yes, pyyaml is a required dependency for building and testing Claude skills, as it supports parsing SKILL.md frontmatter and helper scripts during the benchmarking and packaging workflow.

How does benchmarking work when refining a Claude skill?▼

Benchmarking works by running comparison tests against eval prompts, grading outputs, and spotting flaky or non-discriminating assertions to iteratively refine the SKILL.md content.

Can I use this workflow to update an existing SKILL.md file?▼

Yes, you can use this workflow to revise existing SKILL.md frontmatter and body content, run new benchmark comparison tests, and repackage the updated folder into a distributable file.

Why are my Claude skill eval tests returning flaky results?▼

Flaky eval results often stem from non-discriminating assertions; analyzing benchmark results helps spot these issues so you can refine the SKILL.md triggers based on evidence.

Related Skills

Explore More Software Engineering Skills→
takemo101takemo101

propose-improvement

Analyze workflow improvement ideas, calculate ROI, and create GitHub Issues.

Community
Advanced
Typeless-GitTypeless-Git
1

skill-creator

Automate skill development from intent capture to SKILL.md drafting and evaluations.

Official
Advanced
popiliadampopiliadam
25

schema-validate

Validate JSON Schema Draft7 documents and SKILL.md frontmatter against defined schemas.

Community
Advanced
oabdelmaksoudoabdelmaksoud
4

skill-creator

Draft, test, evaluate, and iterate on OpenClaw SKILL.md-based skills.

Community
Advanced