CourtListener MCP Server
Enables LLM-friendly access to the CourtListener legal database and eCFR for searching legal opinions, court cases, judges, documents, and federal regulations.
README
CourtListener MCP Server
A Model Context Protocol (MCP) server that provides LLM-friendly access to the CourtListener legal database and the Electronic Code of Federal Regulations (eCFR) through the official CourtListener API v4. This server enables searching and retrieving legal opinions, court cases, judges, legal documents, and federal regulations for precise legal research and citation verification.
🎯 Purpose
The CourtListener MCP Server provides comprehensive access to legal case data, court opinions, and federal regulations through the extensive CourtListener and eCFR databases. CourtListener contains millions of legal opinions from federal and state courts, while eCFR provides up-to-date federal regulations.
📋 Key Advantages
- Comprehensive Legal Database:
- Access to millions of court opinions and legal decisions
- Federal and state court coverage
- Real-time updates from court systems
- Full Text Content:
- Complete opinion text for citation verification
- Structured legal document organization
- Rich metadata including judges, courts, and dates
- Regulatory Research:
- Search and retrieve current federal regulations
- Validate regulatory citations and references
- Legal Research:
- Search by judge, court, case name, or content
- Verify exact legal language and precedents
- Validate legal citations and references
🛠️ Available MCP Tools
The CourtListener MCP Server provides these production-ready tools (see app/README.md for full details and parameters):
- Opinion & Case Search:
search_opinions— Search legal opinions and court decisionssearch_dockets— Search court cases and docketssearch_dockets_with_documents— Search dockets with nested documentssearch_recap_documents— Search RECAP filing documentssearch_audio— Search oral argument audiosearch_people— Search judges and legal professionals
- Entity Retrieval:
get_opinion,get_docket,get_audio,get_court,get_person,get_cluster
- Citation & Regulation Tools:
lookup_citation,batch_lookup_citations,verify_citation_format,parse_citation_with_citeurl,extract_citations_from_text,enhanced_citation_lookuplist_titles,list_agencies,search_regulations,list_all_corrections,list_corrections_by_title,get_search_suggestions,get_search_summary,get_title_search_counts,get_daily_search_counts,get_ancestry,get_title_structure,get_source_xml,get_source_json
- System & Health:
status,get_api_status,health_check
See app/README.md for a full reference of all tools, parameters, and usage examples.
📦 Installation
Prerequisites
- Python 3.12+
- uv for dependency management
- Internet connection for CourtListener API access
Install with uv
# Clone the repository
git clone <repository-url>
cd CourtListener
# Install dependencies
uv sync
# Activate the environment (optional)
uv shell
Environment Configuration
Create a .env file in the project root:
COURTLISTENER_BASE_URL=https://www.courtlistener.com/api/rest/v4/
COURT_LISTENER_TIMEOUT=30
LOG_LEVEL=INFO
RATE_LIMIT_REQUESTS=10
RATE_LIMIT_PERIOD=60
DEBUG=false
MCP_PORT=8765
MCP_DEV_PORT=8766
Running the Server
The server now runs with streamable-http transport by default:
uv run python -m app.server
This will start the server at:
- Host:
0.0.0.0(accessible from external connections) - Port:
8000 - Endpoint:
http://localhost:8000/mcp/
Or use the VS Code task: Run MCP Server
Connecting to the Server
When using the streamable-http transport, clients can connect to the server using:
from fastmcp import Client
async with Client("http://localhost:8000/mcp/") as client:
result = await client.call_tool("status")
print(result)
💡 Usage Examples
See app/README.md for detailed tool usage and examples, including search, citation, and regulatory queries.
🐳 Docker Setup
# Production
docker-compose up -d
# Development with hot reload
docker-compose --profile dev up --build
🧪 Testing
uv run pytest
uv run pytest --cov=app --cov-report=term-missing
See tests/README.md for test suite details, coverage, and troubleshooting.
🔧 Development
uv run ruff format .
uv run ruff check .
uv run mypy app/
uv run pip-audit
🚨 Troubleshooting
See app/README.md and tests/README.md for troubleshooting and advanced usage.
📚 Documentation
- Source Code Documentation
- Test Documentation
- Project Context
- CourtListener API Documentation
- eCFR API Documentation
- FastMCP Framework
- Model Context Protocol
Ready to use! The CourtListener MCP Server provides production-ready access to federal regulations and legal data through 20+ comprehensive MCP tools.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。