skill-creator

Create and refine AI skills with SKILL.md metadata and automated testing.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/Witroch4/witdev-skills --skill skill-creator-witroch4
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/Witroch4/witdev-skills/tree/main/skill-creator
Command: npx skills add https://github.com/Witroch4/witdev-skills --skill skill-creator-witroch4

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 skills, from initial concept to performance evaluation, making skill development accessible and efficient.

Core Features & Use Cases

  • Skill Creation: Guides users through defining intent, interviewing for edge cases, and writing SKILL.md files.
  • Iterative Improvement: Facilitates testing, evaluation, and refinement of existing skills based on performance metrics and user feedback.
  • Description Optimization: Automatically tunes skill descriptions for optimal triggering accuracy using a robust evaluation loop.
  • Use Case: You have an idea for a new skill to help Claude summarize meeting notes. Use this Skill to draft the SKILL.md, create test cases, run evaluations, and optimize the description so Claude uses it effectively.

Quick Start

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

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 AI skills for agentic systems?

Build and test AI skills by defining the intent, writing the SKILL.md metadata, bundling resources like scripts, and running automated evaluation loops to measure performance for agentic systems.

What is the process for optimizing AI agent skill descriptions?

Optimizing AI agent skill descriptions involves running an automated evaluation loop that tunes the metadata text to maximize triggering accuracy, ensuring the AI agent selects the correct skill for a given task.

Can I modify and refine existing AI skills based on performance metrics?

Yes, you can iteratively modify and refine existing AI skills by running automated tests, evaluating performance metrics, and updating instructions and bundled references based on the results.

What resources do I need to bundle when creating an AI agent skill?

When creating an AI agent skill, you need to bundle resources such as executable scripts, reference documents, and static assets alongside the SKILL.md file to support the skill's processing logic.

What's the best way to evaluate edge cases during skill development?

The best way to evaluate edge cases during skill development is to use an interviewing process to define them upfront, then generate specific test cases and run automated evaluations to validate the skill's behavior.