skill-creator

Define Skill metadata, test with evals, and iterate on SKILL.md.

1|Updated Apr 1, 2026
One-click install
npx skills add https://github.com/Kingdaddy007/Claude-leaked-source-code --skill skill-creator-kingdaddy007
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/Kingdaddy007/Claude-leaked-source-code/tree/main/stubs/downloads/official-plugins/marketplaces/claude-plugins-official/plugins/skill-creator/skills/skill-creator
Command: npx skills add https://github.com/Kingdaddy007/Claude-leaked-source-code --skill skill-creator-kingdaddy007

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

The Skill Creator provides a structured, repeatable process for turning ideas into self-contained Skill units. It guides you to define discoverable metadata, compose clear instructions, and package optional resources so skills trigger reliably and can be reused across projects.

Core Features & Use Cases

  • Structured Skill modeling: Define name, description, license, compatibility, and metadata to enable discovery and consistent activation.
  • Iterative evaluation loop: Run evals, analyze results, and rewrite SKILL.md to improve triggering accuracy and reliability.
  • Bundled resources management: Organize optional scripts, references, and assets to support advanced behaviors and documentation.

Quick Start

Draft a new skill concept, provide a concise description, and set up an initial SKILL.md with minimal body.

Frequently Asked Questions about skill-creator

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

FAQPage Schema
How do I create a Claude Skill that triggers reliably across different projects?

To create a Claude Skill that triggers reliably, you must define clear discovery metadata, frontmatter, and a concise name within the SKILL.md file to ensure consistent activation across diverse project contexts.

What is the best way to evaluate and refine an existing SKILL.md body?

The best way to evaluate a SKILL.md body is to run structured evals, analyze the test results, and iterate on the description and instructions to systematically improve triggering accuracy and reliability.

How do I bundle scripts and references into a self-contained Skill unit?

You bundle scripts and references into a Skill unit by organizing optional resources like assets and documentation alongside the SKILL.md file to support advanced behaviors and ensure reproducible outcomes.

Does the Skill Creator workflow require specific dependencies like anthropic and pyyaml?

Yes, the Skill Creator workflow requires the anthropic and pyyaml dependencies to properly structure metadata, format configuration files, and execute the iterative evaluation loop during skill creation.

What limitations exist when optimizing workflow creation benchmarks for Claude Skills?

Limitations in workflow optimization arise when discovery metadata is poorly defined, causing unreliable skill triggering; continuous evaluation and rewriting of the SKILL.md body are required to overcome these activation constraints.