LLM Docs Optimizer
A Claude Code skill that makes your documentation work seamlessly with AI coding assistants. Transform generic docs into AI-optimized documentation with measurable quality improvements—scored by Context7's c7score benchmark, structured around developer questions, and enhanced with LLM-friendly navigation.
Why Use This?
The Problem:
Standard documentation often fails with AI coding assistants. Generic API references, incomplete examples, and poor structure make it hard for tools like Claude and GitHub Copilot to help developers effectively.
The Solution:
This skill transforms your documentation to work seamlessly with AI tools by:
- Measurable Quality: Score your docs against Context7's c7score benchmark and track improvements (e.g., 45 → 89/100)
- Better AI Retrieval: Organize content around developer questions so LLMs can find the right answers
- Complete Examples: Transform API references into runnable, practical code snippets
- LLM-Friendly Navigation: Generate standardized llms.txt files that help AI understand your project structure
- Dual Benefits: Better for human developers AND AI assistants
Real Impact:
- Developers get answers faster when using AI coding assistants
- Your documentation becomes more practical and question-driven
- AI tools can better understand and recommend your project
- Measurable before/after quality metrics prove the improvement
Demo video
https://github.com/user-attachments/assets/3ad0ff3f-69bc-4553-bbfd-654b0880e3b5
Overview
This skill provides comprehensive documentation optimization for AI tools:
- C7Score Optimization: Transform documentation to score highly on Context7's benchmark - the leading quality metric for AI-assisted coding documentation
- llms.txt Generation: Create standardized navigation files that help LLMs quickly understand and navigate your project's documentation
- Automated Quality Scoring: Get before/after evaluation across 5 key metrics to measure improvement
- Question-Driven Restructuring: Organize content around developer questions for better AI retrieval
What is c7score?
c7score is a benchmark created by Context7 that measures documentation quality for AI-assisted coding. It evaluates:
- Question-Snippet Matching (80%): How well code snippets answer common developer questions
- LLM Evaluation (10%): Overall comprehensiveness and clarity
- Formatting (5%): Proper markdown structure and code block formatting
- Metadata Removal (2.5%): Elimination of irrelevant timestamps and badges
- Initialization Examples (2.5%): Quality of setup and quickstart examples
What is llms.txt?
llms.txt is a standardized markdown file format from llmstxt.org that provides LLM-friendly documentation navigation. It helps AI tools quickly understand your project structure and find relevant information.
Features
C7Score Optimization
- 📊 Documentation Analysis: Automatically identifies quality issues and gaps
- ✨ Smart Optimization: Applies proven patterns to improve LLM retrieval
- 📝 Code Snippet Enhancement: Transforms examples to better answer developer questions
- 🎯 Question-Driven Structure: Organizes content around common developer queries
- 📈 Automated Scoring: Claude-based evaluation across all 5 c7score metrics with before/after comparison
- 🔍 Python Analysis Tool: Automated script to scan documentation for issues
- 📚 Reference Patterns: Library of transformation examples and best practices
llms.txt Generation
- 🗺️ Project Analysis: Understands your documentation structure
- 📋 Smart Templates: Uses appropriate format for your project type (library, CLI, framework, etc.)
- 🔗 Proper Formatting: Follows official llmstxt.org specification
- 📝 Complete Examples: Includes before/after examples and templates
- 🎯 Priority Organization: Structures content from essential to optional
Installation
Quick Install (Recommended)
Install directly from the marketplace using Claude Code:
# Step 1: Add the marketplace (one-time setup)
/plugin marketplace add alonw0/llm-docs-optimizer
# Step 2: Install the plugin
/plugin install llm-docs-optimizer@llm-docs-optimizer-marketplace
The plugin will be automatically available in Claude Code after installation.
Alternative: Manual Installation
If you prefer manual installation:
- Clone the repository:
git clone https://github.com/alonw0/llm-docs-optimizer.git ~/.claude/plugins/llm-docs-optimizer