skills.rest
riethmayer
riethmayerCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

skill-creator

Create, evaluate, and refine Claude Skills through iterative SKILL.md drafting and eval loops.

2|1|Updated Oct 8, 2009
Search Tags:#workflow#evaluation#iteration#claude-code#skill-creation#prompt-engineering
One-click install
npx skills add https://github.com/riethmayer/dotfiles --skill skill-creator-riethmayer
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/riethmayer/dotfiles/tree/main/stow/agents/.agents/skills/skill-creator
Command: npx skills add https://github.com/riethmayer/dotfiles --skill skill-creator-riethmayer

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The Skill Creator helps teams and individuals design, test, and iteratively improve Claude skills. It guides users through capturing intent, researching best practices, drafting SKILL.md, and running evals to measure performance, then repeating until triggers are reliable and outputs meet user goals.

Core Features & Use Cases

  • Capture Intent: define what a skill should enable Claude to do and when to trigger.
  • Iterative Improvement: run evaluation cycles, analyze results, and generate improved SKILL.md and descriptions.
  • Automation Toolkit: manage test prompts, evals, and history to accelerate skill maturation.
  • Use Case: A product manager wants to craft a new skill to automate a user onboarding task, then evaluate and refine it until triggering is robust across variations.

Quick Start

Use the skill-creator to outline a new skill, create a draft SKILL.md, and kick off an eval loop to test trigger accuracy and refine the description based on results.

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 refine Claude skills using an eval-driven loop?▼

To create and refine Claude skills, you capture intent, draft a SKILL.md file, run evaluation prompts, analyze results, and iterate on descriptions and tooling to maximize triggering accuracy and useful outputs.

What is the best way to evaluate and improve prompt triggering accuracy?▼

The best way to improve prompt triggering accuracy is to run iterative evaluation cycles, analyze test prompt results, and generate improved SKILL.md descriptions based on those outcomes.

How do I draft a SKILL.md file for a new workflow automation task?▼

You draft a SKILL.md file by first capturing the intent of your workflow automation task, researching best practices, and outlining what the skill should enable Claude to do and when to trigger it.

Can I use Claude Code to automate skill creation and testing for my team?▼

Yes, you can use Claude Code to automate skill creation and testing for your team by managing test prompts, running evals, and tracking history to accelerate skill maturation across variations.

Does the skill-creator workflow require external dependencies to run evals?▼

No, the skill-creator workflow does not require external dependencies to run evals; it operates independently to manage test prompts, analyze results, and refine SKILL.md files.

Why does my Claude skill fail to trigger reliably across different prompt variations?▼

Your Claude skill fails to trigger reliably because the initial SKILL.md description lacks sufficient refinement; you must run eval prompts, analyze failures, and iterate on the description to fix triggering gaps.

Related Skills

Explore More Content & Communication Skills→
8b-is8b-is

creating-agent-skills

Design and refine Claude Code Skills with consistent SKILL.md structure.

Official
Advanced
CHENyiru3CHENyiru3

skill-creator

Automate end-to-end creation, evaluation, and refinement of Claude Code skills.

Community
Advanced
wsxwj123wsxwj123
7

skill-creator

Draft Claude skills, evaluate prompts, and iterate on trigger descriptions.

Community
Advanced
yahav123147yahav123147

skill-creator

Create and iterate SKILL.md structures with trigger evaluations and grading evidence.

Community
Advanced