legal-mcp
Enables US case law search, citation parsing, practice management via Clio, and federal court filings through PACER.
README
Legal MCP Server
US case law search, citation parsing, practice management, and federal court filings for AI assistants — powered by FastMCP.
Hosted at https://legal-mcp-production.up.railway.app/mcp
Setup
Add the server to your Claude Code settings (~/.claude/settings.json):
{
"mcpServers": {
"legal-mcp": {
"type": "url",
"url": "https://legal-mcp-production.up.railway.app/mcp",
"headers": {
"X-CourtListener-Token": "YOUR_COURTLISTENER_TOKEN"
}
}
}
}
Getting a CourtListener token
- Create a free account at https://www.courtlistener.com/sign-in/
- Go to your profile and copy your API token
- Add it as the
X-CourtListener-Tokenheader above
Optional: Clio practice management
If you have a Clio account with API access, add your OAuth token:
"headers": {
"X-CourtListener-Token": "YOUR_COURTLISTENER_TOKEN",
"X-Clio-Token": "YOUR_CLIO_OAUTH_TOKEN"
}
Available tools
| Tool | Description |
|---|---|
search_case_law |
Search 4M+ US court opinions by topic, court, or date range |
get_case_details |
Get full text and details of a specific opinion |
get_case_record |
Get docket info — parties, attorneys, judges, procedural history |
find_citing_cases |
Find cases that cite a given opinion |
find_cited_cases |
Find cases cited by a given opinion |
parse_legal_citations |
Parse Bluebook-format citations from text |
list_available_courts |
List all US courts and their filter codes |
list_reporter_abbreviations |
Reference for reporter abbreviations (U.S., F.3d, etc.) |
search_federal_cases |
Search PACER for federal cases (requires PACER credentials) |
get_federal_case |
Get federal case details from PACER |
get_court_filings |
Get docket entries for a federal case from PACER |
search_clients |
Search Clio contacts (requires Clio token) |
search_matters |
Search Clio matters/cases |
get_matter_details |
Get full matter details from Clio |
get_time_entries |
Get billable hours from Clio |
get_matter_tasks |
Get tasks for a matter in Clio |
get_matter_documents |
Search documents in Clio |
get_calendar |
Get calendar entries from Clio |
Self-hosting
git clone https://github.com/jscottym/legal-mcp.git
cd legal-mcp
pip install -r requirements.txt
python -m legal_mcp.server
Or with Docker:
docker build -t legal-mcp .
docker run -p 8000:8000 -e COURTLISTENER_TOKEN=your_token legal-mcp
Environment variables
| Variable | Description |
|---|---|
PORT |
Server port (default: 8000) |
COURTLISTENER_TOKEN |
Fallback CourtListener API token |
CLIO_TOKEN |
Fallback Clio OAuth token |
PACER_USERNAME |
PACER login |
PACER_PASSWORD |
PACER password |
LEGAL_MCP_DEMO |
Set to true to use demo data without API keys |
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。