@courtmesh/mcp-server

An MCP (Model Context Protocol) server for the CourtMesh public REST API. It gives any MCP capable AI client, Claude Desktop, Claude Code, Cursor, or a custom agent, tools to search and analyze Indian court case law: 310M plus case records spanning the Supreme Court, High Courts, District Courts and tribunals.

Built with the official @modelcontextprotocol/sdk. MIT licensed.

Quickstart

1. Get an API key

Sign up and generate a key at https://research.courtmesh.ai. Keys look like cm-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx-xxxx (a legacy vv- prefix also works).

2. Run it

You do not need to install anything by hand, npx will fetch and run the package. Set COURTMESH_API_KEY in your MCP client config, see the copy pasteable blocks below.

Local checkout:

npm install
npm run build
COURTMESH_API_KEY=cm-your-key-here node dist/index.js

Related MCP server: canlii-mcp

Environment variables

Transports

This server supports two transports, chosen at startup:

• stdio (default): the standard transport for local MCP clients such as Claude Desktop, Claude Code and Cursor. The client spawns the server process and talks to it over stdin and stdout.

• Streamable HTTP: pass --http or set MCP_TRANSPORT=http to run as a long lived HTTP server, mounted at /mcp, suitable for hosting behind a URL such as mcp.courtmesh.ai/mcp. A plain GET /health route is also available on the HTTP server itself, separate from the check_api_health tool, which checks the upstream CourtMesh API instead.

Tools

Full input and output field details are in each tool's own description, visible to any connected MCP client through tools/list.

Configuration examples

Claude Desktop

Edit claude_desktop_config.json (Settings, Developer, Edit Config) and add:

{
  "mcpServers": {
    "courtmesh": {
      "command": "npx",
      "args": ["-y", "@courtmesh/mcp-server"],
      "env": {
        "COURTMESH_API_KEY": "cm-your-key-here"
      }
    }
  }
}

Claude Code

CLI one liner:

claude mcp add courtmesh --env COURTMESH_API_KEY=cm-your-key-here -- npx -y @courtmesh/mcp-server

Or add to .mcp.json in your project:

{
  "mcpServers": {
    "courtmesh": {
      "command": "npx",
      "args": ["-y", "@courtmesh/mcp-server"],
      "env": {
        "COURTMESH_API_KEY": "cm-your-key-here"
      }
    }
  }
}

Cursor

Edit ~/.cursor/mcp.json:

{
  "mcpServers": {
    "courtmesh": {
      "command": "npx",
      "args": ["-y", "@courtmesh/mcp-server"],
      "env": {
        "COURTMESH_API_KEY": "cm-your-key-here"
      }
    }
  }
}

Hosted HTTP mode

Once this server is deployed behind a public URL, point any Streamable HTTP capable MCP client at it directly, no local process required. Pass your key as a query parameter, it overrides COURTMESH_API_KEY for that session:

{
  "mcpServers": {
    "courtmesh": {
      "url": "https://mcp.courtmesh.ai/mcp?token=cm-your-key-here"
    }
  }
}

Error codes

The API's own error text is always surfaced verbatim where available, along with a plain explanation.

There is one API quirk this server handles for you: POST /search/cases/semantic sends its HTTP 200 status before it finishes work, so a failure inside that endpoint can still arrive as HTTP 200 with a body of {"success": false, "error": "..."}. Every tool call checks for success: false in the response body in addition to the HTTP status, and reports it as a tool error either way.

Development

npm install
npm run build   # compiles TypeScript with tsc, then chmods dist/index.js executable
npm run dev      # tsc --watch
npm start        # runs the built server over stdio

Source layout:

• src/index.ts: entry point, transport selection (stdio vs Streamable HTTP), Express app for HTTP mode.

• src/client.ts: shared HTTP client, auth headers, and all error mapping.

• src/tools.ts: tool schemas (zod) and handlers, one per CourtMesh endpoint.

• src/context.ts: an AsyncLocalStorage used to carry a per request ?token= override through to the client in HTTP mode.

License

MIT, Copyright 2026 Thinkscoop Technologies LLP. See LICENSE.