courtlistener-mcp

CourtListener MCP Integration

AI-powered legal research using CourtListener API with Model Context Protocol (MCP) integration for Claude

Project Overview

A comprehensive legal research tool that combines:

• MCP Server: Enables Claude to conduct legal research via CourtListener API
• Judicial Writing Analysis: AI model training for judge-specific opinion generation
• Citation Verification: Bluebook citation validation and formatting

🚀 Features

MCP Integration (In Development)

• Case Law Search: Full-text search across millions of court opinions
• Citation Verification: Validate case citations and Bluebook formatting
• Docket Access: Retrieve detailed docket information
• Smart Filtering: Filter by jurisdiction, date, and precedential status
• Natural Language: Claude interprets natural language legal queries

Judicial Writing Analysis (Complete ✅)

• Judge Style Profiles: Analyze individual judicial writing patterns
• Training Data Generation: Create AI training datasets from real opinions
• Citation Pattern Analysis: Extract citation integration styles
• Bluebook Compliance: Ensure proper legal citation formatting

📁 Repository Structure

courtlistener-mcp/
├── mcp-server/           # MCP server implementation
│   ├── index.js          # Main server entry point
│   └── tools/            # CourtListener API tools
├── judge-analysis/       # Judicial writing analysis
│   ├── comprehensive_foley_training.txt
│   └── style-analyzer.js
├── docs/                 # Documentation
└── examples/             # Usage examples

🛠️ Installation

# Clone the repository
git clone https://github.com/open-legal-tools/courtlistener-mcp.git
cd courtlistener-mcp

# Install dependencies
npm install

# Configure API credentials
cp .env.example .env
# Edit .env with your CourtListener API key

📖 Usage

MCP Server with Claude Desktop

1. Add to your Claude desktop configuration:

{
  "mcpServers": {
    "courtlistener": {
      "command": "node",
      "args": ["path/to/courtlistener-mcp/mcp-server/index.js"],
      "env": {
        "COURTLISTENER_API_KEY": "your-api-key"
      }
    }
  }
}

2. Use natural language queries:
   • "Find recent Supreme Court cases about AI and copyright"
   • "Check if Miranda v. Arizona is still good law"
   • "Show me 9th Circuit cases on qualified immunity from 2023"

Judge Writing Analysis

# Analyze a judge's writing style
node judge-analysis/style-analyzer.js --judge "Foley" --court "ind"

# Generate training data
node judge-analysis/training-generator.js --input opinions/ --output training.txt

🔧 API Reference

Available MCP Tools

search_opinions

{
  query: string;          // Search query
  court?: string;         // Court identifier
  after?: string;         // Date after (YYYY-MM-DD)
  before?: string;        // Date before (YYYY-MM-DD)
  precedential?: boolean; // Only precedential opinions
}

verify_citation

{
  citation: string;  // Full citation string
}

get_citing_cases

{
  id: string;     // Opinion ID
  limit?: number; // Maximum results
}

🎯 Current Status

• Phase 1 Complete ✅: Judge writing style analysis and training data
• Phase 2 In Progress 🔄: MCP server implementation
• Phase 3 Planned 📋: Advanced citation network analysis

🤝 Contributing

Contributions welcome! See CONTRIBUTING.md for guidelines.

📄 License

MIT License - see LICENSE for details.

🙏 Acknowledgments

• CourtListener by the Free Law Project
• Model Context Protocol by Anthropic
• Indiana Court of Appeals for publicly available opinions

📞 Support

• Issues: GitHub Issues
• CourtListener API: Documentation
• MCP Docs: Anthropic MCP Guide