CodeDreamer06/BookDownloaderMCP

LibGen Downloader + MCP Server

A powerful Node.js/TypeScript tool for searching and downloading ebooks from LibGen mirrors, featuring a Model Context Protocol (MCP) server for seamless LLM integration.

🚀 Features

• MCP Server Integration: Enable LLMs (like Claude) to search and download books via tool calls
• Programmatic API: Full Node.js API for automated workflows
• Bulk Downloads: Download multiple books simultaneously by MD5
• Smart Mirror Discovery: Automatic detection and failover for LibGen+ mirrors
• Resilient Operations: Built-in retries and error handling for network requests
• TypeScript Support: Full type definitions included

📦 Installation

For MCP Server (LLM Integration)

npm install -g book-downloader-mcp

For Programmatic Use

npm install book-downloader-mcp

Requirements: Node.js 16+ recommended

🤖 MCP Server Setup

Claude Desktop Integration

1. Install globally:

npm install -g book-downloader-mcp

2. Add to your claude_desktop_config.json:

{
  "mcpServers": {
    "libgen": {
      "command": "libgen-mcp"
    }
  }
}

3. Restart Claude Desktop

The assistant will now have access to LibGen search and download capabilities!

Other MCP Clients

Run the MCP server directly:

libgen-mcp

Or from source:

npm run start:mcp

💻 Programmatic Usage

Basic Example

const libgen = require('@rovodev/libgen-mcp-server');

async function searchAndDownload() {
  // Initialize mirror connection
  await libgen.ensureMirrorReady();
  
  // Search for books
  const results = await libgen.searchBooks('The Art of War');
  console.log(`Found ${results.length} results`);
  
  if (results.length > 0) {
    const book = results[0];
    console.log(`Downloading: ${book.title} by ${book.authors}`);
    
    // Get direct download URL
    const url = await libgen.getDirectDownloadUrlByMd5(book.md5);
    console.log(`Download URL: ${url}`);
    
    // Download the file
    const outcome = await libgen.downloadByMd5(book.md5);
    if (outcome.status === 'ok') {
      console.log(`Downloaded to: ${outcome.path}`);
    }
  }
}

searchAndDownload();

TypeScript Example

import * as libgen from '@rovodev/libgen-mcp-server';

async function bulkDownload(searchQuery: string) {
  await libgen.ensureMirrorReady();
  
  const results = await libgen.searchBooks(searchQuery, 1, 10);
  const md5List = results
    .filter(book => book.md5)
    .map(book => book.md5!)
    .slice(0, 3); // Download first 3 books
  
  const outcomes = await libgen.downloadByMd5List(md5List);
  outcomes.forEach(outcome => {
    if (outcome.status === 'ok') {
      console.log(`✓ Downloaded: ${outcome.filename}`);
    } else {
      console.log(`✗ Failed: ${outcome.error}`);
    }
  });
}

API Reference

• ensureMirrorReady(): Initialize mirror configuration
• searchBooks(query, page?, pageSize?): Search for books
• getDirectDownloadUrlByMd5(md5): Get download URL for a book
• downloadByMd5(md5): Download a single book
• downloadByMd5List(md5Array): Bulk download multiple books

🛠 MCP Tools Reference

When connected via MCP, the following tools are available to LLMs:

Core Operations

Search & Discovery

Search Result Fields: id, authors, title, publisher, year, pages, language, size, extension, mirror, md5, resolvedMirrorUrl

Download Operations

Example MCP Workflow

// 1. Initialize
await mcp.call("libgen.ensureMirror", {});

// 2. Search
const results = await mcp.call("libgen.search", { 
  query: "machine learning", 
  page: 1 
});

// 3. Download first result
if (results.length > 0) {
  const outcome = await mcp.call("libgen.download", { 
    md5: results[0].md5 
  });
}

⚠️ Important Notes

• Mirror Status: LibGen mirrors can be unstable. The tool automatically discovers working mirrors from the configuration
• File Locations: Downloads are saved to the current working directory with original filenames
• Network Resilience: Built-in retries handle temporary network issues
• LibGen+ Compatibility: Optimized for LibGen+ mirror structure and parsing

🔧 Development

Building the Project

npm run build          # Compile TypeScript
npm run build:mcp      # Build and run MCP server
npm run watch          # Watch mode for development

Running Locally

npm run start:mcp      # Start MCP server in development
npm run lint           # Run ESLint
npm run format         # Format code with Prettier

Testing

npm test               # Run all tests
npm run test:ci        # Run tests in CI mode

Optional Integration Tests: Set environment variables for Groq API testing:

• GROQ_API_KEY: Your Groq API key
• RUN_GROQ_TESTS=1: Enable integration tests

Create a .env file (gitignored) or use .env.test for testing. See .env.example for reference.

🏗 Architecture

• Core Layer: HTML parsing (jsdom), HTTP requests, mirror discovery, retry logic
• MCP Server: Stdio-based server exposing LibGen operations as MCP tools
• Programmatic API: Direct function exports for Node.js integration
• Mirror Management: Dynamic configuration fetching and automatic failover

📋 Changelog

v3.0.2 (Current)

• ✅ Full test coverage with proper mocking
• ✅ Updated MCP SDK compatibility
• ✅ Enhanced error handling and retries
• ✅ Improved documentation and examples

v3.0.0

• 🔄 LibGen+ mirrors as primary source
• ➖ Removed search filtering for LibGen+ compatibility
• ➖ Temporarily removed alternative downloads

v2.x

• ➕ Alternative downloads, progress indicators
• ➕ CLI search capabilities and shortcuts
• ➕ Caching system

v1.x

• 🔄 Complete UI rewrite with React/Ink/Zustand
• ➕ Bulk download functionality
• ✅ Enhanced error handling and retry logic

📄 License

WTFPL - Do What The F*ck You Want To Public License

⚖️ Legal Disclaimer

This tool accesses third-party LibGen mirrors which may be:

• Blocked or unstable in certain regions
• Subject to local laws and regulations

Use responsibly and ensure compliance with applicable laws in your jurisdiction.

🤝 Contributing

1. Fork the repository
2. Create a feature branch
3. Add tests for new functionality
4. Ensure all tests pass: npm test
5. Submit a pull request

Repository: https://github.com/obsfx/libgen-downloader