context-extractor

Extract project conventions and produce a CLAUDE.md artifact from repository contents.

Updated Feb 19, 2026
One-click install
npx skills add https://github.com/kzarzycki/agent-skills --skill context-extractor-kzarzycki
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: context-extractor
Source: https://github.com/kzarzycki/agent-skills/tree/main/engineering/skills/context-extractor
Command: npx skills add https://github.com/kzarzycki/agent-skills --skill context-extractor-kzarzycki

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

It prevents onboarding and execution mistakes by capturing the conventions, patterns, and project-specific context that an AI might otherwise infer incorrectly.

Core Features & Use Cases

  • Project-wide conventions extraction: Detects project type(s) and scans for structure, naming patterns, and existing documentation signals.
  • Type-aware checklist scanning: Tailors the conventions search to code repos, CI pipelines, presentations, documentation sites, and data/research notebooks.
  • Actionable CLAUDE.md drafting: Produces a concise CLAUDE.md (or proposed additions) that Claude can follow for build/test/run and “gotchas”.
  • Safe update behavior: Never removes existing CLAUDE.md content; it proposes additions only when a file already exists.

Quick Start

Ask your AI to run context extraction on the current repo and generate a CLAUDE.md focused on commands, conventions, and gotchas.

Frequently Asked Questions about context-extractor

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

FAQPage Schema
How do I generate a CLAUDE.md file for my project to capture build commands and gotchas?

Generating a CLAUDE.md playbook involves scanning the repository to detect project type, structure, and existing documentation, then drafting an artifact that covers build commands, conventions, and gotchas for reliable onboarding and execution.

What is a CLAUDE.md playbook and why do I need one for onboarding?

A CLAUDE.md playbook is a contextual guidance file that captures your project's conventions, patterns, and execution rules. You need it to prevent onboarding and execution mistakes by ensuring AI understands project-specific context rather than inferring it incorrectly.

Can I extract project conventions for data research notebooks and CI pipelines?

Yes, context extraction applies type-aware checklist scanning tailored to code repos, CI pipelines, documentation sites, presentations, and data/research projects to capture type-specific conventions and patterns.

Does generating a CLAUDE.md playbook overwrite my existing project documentation?

No, the extraction features safe update behavior that never removes existing user-authored content. When a CLAUDE.md file already exists, it only proposes additions to preserve your original project documentation and conventions.

What's the best way to document project-specific patterns and formatting configs for AI execution?

The best way is to run context extraction, which scans your repository for organization, naming patterns, and formatting configs, then drafts a CLAUDE.md artifact that Claude can follow for build, test, and run execution.