skill-creator

Automate the end-to-end workflow of defining, evaluating, and improving Claude skills.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/Wakewell-Sleep-Solutions/claude-skills-ecosystem --skill skill-creator-wakewell-sleep-solutions
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/Wakewell-Sleep-Solutions/claude-skills-ecosystem/tree/main/skill-creator
Command: npx skills add https://github.com/Wakewell-Sleep-Solutions/claude-skills-ecosystem --skill skill-creator-wakewell-sleep-solutions

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

The Skill Creator addresses the challenge of turning user intents into robust, testable Claude skills. It guides you from initial draft to evaluable skill by structuring SKILL.md, outlining objectives, and modeling an iterative loop that evaluates, compares, and improves skill performance.

Core Features & Use Cases

  • Capture Intent and Scope: document what the skill should enable and how it should trigger.
  • Iterative Evaluation: define evals, run trigger tests, review results with an integrated viewer, and iterate on the skill description and implementation.
  • Packaging and Reuse: package finished skills into distributable bundles and share across teams.
  • Use Case: a product team wants to draft a new skill, validate triggering, refine the description for reliability, and prepare a reusable package for deployment.

Quick Start

Draft a SKILL.md, define a small eval set, run the end-to-end loop, review results, and iteratively improve the skill before packaging.

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 evaluate Claude skills iteratively?

To create and evaluate Claude skills, draft a SKILL.md, define eval prompts, run trigger tests, review results with an integrated viewer, and iterate on the description and implementation before packaging.

What is the best way to package Claude skills for team distribution?

The best way to package Claude skills is to use an automated workflow that bundles finished, evaluated skills into distributable packages, ensuring reliable triggering and implementation before sharing across teams.

How do I test if a Claude skill triggers correctly for specific intents?

To test if a Claude skill triggers correctly, define eval prompts, run trigger tests within an iterative evaluation loop, and use the integrated viewer to review results and refine the skill scope.

Can I use Anthropic and PyYAML to automate skill creation and benchmarking?

Yes, you can use Anthropic and PyYAML dependencies to automate the end-to-end workflow of defining, evaluating, and benchmarking Claude skills, including drafting SKILL.md and running trigger tests.

What should I include in a SKILL.md file to ensure reliable skill performance?

A SKILL.md file should include the skill's objectives, intended scope, and trigger conditions to ensure reliable performance, which you then validate through iterative evaluation and trigger tests.

Why does my Claude skill trigger incorrectly during evaluation?

A Claude skill triggers incorrectly when its description and scope lack precision, which you can resolve by running trigger tests, reviewing eval results, and iterating on the SKILL.md implementation.