Skip to Content

Claude Chunks

View original on GitHub 

An intelligent document chunking MCP server optimized for Claude context windows. Split large documents into meaningful, context-aware chunks that can be progressively processed by Claude.

Overview

Claude Chunks helps you process large documents (like books, theses, or long papers) by:

  1. Breaking them into meaningful sections
  2. Generating rich summaries of each section
  3. Maintaining context and connections between sections
  4. Formatting output for optimal Claude context reuse

Installation

# Clone the repository git clone https://github.com/vetlefo/claude-chunks.git cd claude-chunks # Install dependencies npm install # Build the project npm run build

Usage with Claude Desktop

  1. Add to Claude Desktop config:
{ "mcpServers": { "claude-chunks": { "command": "node", "args": ["/path/to/claude-chunks/dist/index.js"] } } }
  1. Restart Claude Desktop
  2. Use the chunk_document tool in your conversations

Development Roadmap

Phase 1: Core Functionality

  • Basic MCP server setup
  • Smart document chunking
  • Section summarization
  • Context preservation

Phase 2: Enhancements

  • Multiple document formats
  • Custom chunking strategies
  • Enhanced metadata tracking
  • Claude-optimized formatting

Phase 3: Advanced Features

  • Cross-reference tracking
  • Technical term extraction
  • Theme detection
  • Progressive loading

Contributing

Contributions are welcome! Please see our Contributing Guide for details.

License

MIT

Last updated on