skill-extractor

Extract reusable scripts, references, and assets from Claude Code commands into modular skills.

40|7|Updated Oct 20, 2025
One-click install
npx skills add https://github.com/rawveg/skillsforge-marketplace --skill skill-extractor
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-extractor
Source: https://github.com/rawveg/skillsforge-marketplace/tree/main/skill-extractor
Command: npx skills add https://github.com/rawveg/skillsforge-marketplace --skill skill-extractor

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill addresses the challenge of managing large, complex AI commands and agents by automatically identifying and extracting reusable components (scripts, references, assets) into modular, standalone skills. This saves development time, improves code maintainability, and promotes reusability across your AI ecosystem.

Core Features & Use Cases

  • Automated Component Analysis: Scans existing commands/agents to detect large code blocks, documentation, and templates suitable for extraction.
  • Structured Skill Generation: Creates properly formatted skill directories with SKILL.md, scripts/, references/, and assets/ as needed.
  • Detailed Migration Reports: Provides step-by-step guidance on how to update your original command/agent to reference the newly extracted skill, ensuring a smooth transition.
  • Use Case: You have a "Content Creator" agent with embedded writing style guides, boilerplate article templates, and a Python script for SEO keyword analysis. Use this skill to extract these into a "Writing Tools" skill, making them accessible to other content-related agents and simplifying your main agent's logic.

Quick Start

To analyze and extract components from an existing command: Analyze a command: @skill-extractor analyze --input ~/.claude/commands/mine/my-command.md Extract components: @skill-extractor extract analysis.json --name my-new-skill --output ./my-skills

Frequently Asked Questions about skill-extractor

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

FAQPage Schema
How do I extract reusable components from large Claude Code commands and agents?

Skill Extractor automatically scans your commands and agents to identify and extract reusable components—scripts (≥20 lines), references (≥500 words), and assets (≥200 characters)—into modular, standalone skills with structured SKILL.md directories, migration reports, and distributable packages.

Can I modularize embedded templates and scripts from my existing agents?

Yes. Skill Extractor detects embedded templates, references, scripts, and assets within monolithic agents, extracts them into properly formatted skill directories, and generates step-by-step migration guidance to update your original agent to reference the new skill.

What's the fastest way to turn complex agent logic into reusable skills?

Run the analyze command on your existing agent to detect extractable components, then use the extract command with your chosen skill name and output directory. Skill Extractor generates SKILL.md, component directories, validation, and a detailed migration report automatically.

Do I need Python installed to use Skill Extractor?

Yes. Skill Extractor requires PyYAML, installed via `pip install pyyaml`. This dependency enables structured component analysis and metadata parsing for your extraction workflows.

How do I know which parts of my command are worth extracting into a separate skill?

Skill Extractor enforces pattern-based detection thresholds: scripts must be ≥20 lines, references ≥500 words, and assets ≥200 characters. The analyze command identifies components meeting these criteria and flags them as extraction candidates in your migration report.

Can Skill Extractor work with commands that use multiple asset types?

Yes. Skill Extractor handles mixed component types—code scripts, documentation references, and template assets—within a single command. It structures extracted skills with separate `scripts/`, `references/`, and `assets/` directories as needed, then validates the package for distribution.