skill-harvester

Extract reusable Claude Code skills from git history and generate Markdown skill templates.

16|7|Updated Nov 20, 2025
One-click install
npx skills add https://github.com/jackspace/ClaudeSkillz --skill skill-harvester
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-harvester
Source: https://github.com/jackspace/ClaudeSkillz/tree/main/skills/skill-harvester
Command: npx skills add https://github.com/jackspace/ClaudeSkillz --skill skill-harvester

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This meta-skill analyzes past work sessions to extract reusable Claude Code skills, then generates comprehensive skill definitions with templates and best practices.

Core Features & Use Cases

  • Analyze git history, code patterns, workflows
  • Identify repeatable patterns and workflows
  • Generate well-structured skill definitions with templates
  • Save best practices and examples for sharing
  • Create organization-wide knowledge repositories

Quick Start

Run the harvest process on a completed project to generate a new Claude Skill you can publish.

Frequently Asked Questions about skill-harvester

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

FAQPage Schema
How do I extract reusable skills from completed projects and past work?

Skill harvesting analyzes git history, code patterns, and documentation to identify repetitive workflows and extract them into structured Claude Code skill definitions. Run the harvest process on a finished project to generate a shareable skill with metadata, templates, and best practices ready for your organization's knowledge base.

Can I build an organization-wide skill library from my team's workflows?

Yes. The harvester generates directory-organized skill definitions with tagging and licensing that populate shareable repositories. Extract patterns from multiple projects, tag them by domain—automation, backend, DevOps, security, frontend, documentation—and structure output for vector databases and RAG pipelines for team discovery and reuse.

What patterns does skill harvesting identify from code and git history?

The skill analyzes git commits, code structure, and workflow documentation to recognize repetitive tasks, automation sequences, and domain-specific processes. It generates Markdown templates with metadata extraction, flagging harvestable patterns in scenarios like project completion, recurring manual steps, and complex documented workflows suitable for automation.

How do extracted skills integrate with RAG pipelines and vector databases?

Harvested skills output in formats compatible with vector databases and retrieval-augmented generation systems. Structured metadata, Markdown templates, and organized skill definitions enable indexing and semantic search, allowing your organization to retrieve and reuse relevant skills across automation, data engineering, and DevOps contexts.

What's the difference between harvesting skills from documentation versus code?

Harvesting analyzes both. Code-based extraction identifies automation patterns and technical workflows from repositories; documentation analysis captures procedural knowledge and best practices. Combined extraction from git history, source files, and docs creates comprehensive skill definitions covering implementation details, prerequisites, and usage guidance for broader reuse.

Do I need existing Claude Code skills to start harvesting from my work?

No prerequisites required. Start with any completed project, codebase, git history, or documentation. The harvester identifies extractable patterns and generates new skill definitions from scratch, building your library incrementally as you run the harvest process on different projects and workflows across your organization.