StatPearls

You run a StatPearls MCP Server to fetch disease information from StatPearls and provide well-formatted, Markdown-friendly content to AI models through the Model Context Protocol (MCP). This server makes medical data easily searchable, structured, and ready for AI consumption.

How to use

Start by launching the MCP server on your machine using one of the supported local runtime methods. The server exposes a single tool, statpearls_disease_info, which you can call through your MCP client to retrieve comprehensive disease information. The output is formatted in Markdown, including a title, optional table of contents, and structured sections such as etiology, epidemiology, clinical features, diagnosis, treatment, and prognosis when available.

How to install

Prerequisites you need before installing:

• A MCP-capable AI client (e.g., Claude Desktop or a compatible MCP client).
• A runtime environment for your chosen installation method (Windows, macOS, Linux, or Node.js for building from source).

Install from binaries (most straightforward)

# For Windows (standalone executable)
# Place the executable wherever you want, then reference it in your MCP settings:
{
  "mcpServers": {
    "statpearls": {
      "command": "{path_to_executable_here}\\statpearls-mcp.exe"
    }
  }
}

Alternatively, for macOS or Linux, use the appropriate Linux/macOS binary location in your MCP settings.

Install from source (advanced)

# Start from a clean clone
# Ensure Bun is installed on your system
clone_url_here/statpearls-mcp
cd statpearls-mcp
bun install
bun run build

# Add the server to your mcp_settings.json
{
  "mcpServers": {
    "statpearls": {
      "command": "node",
      "args": [
        "{path_to_proj_here}/dist/index.js"
      ]
    }
  }
}

Via Smithery (optional auto-install)

You can automatically install statpearls-mcp for Claude Desktop using Smithery. Run the following command in your terminal to install the MCP server for Claude as a client.

# Install as a Claude client via Smithery
npx -y @smithery/cli install @jpoles1/statpearls-mcp --client claude

Configuration and startup notes

If you are using the binaries, you generally only need to point your MCP client to the local executable via the settings shown in the installation steps. If you build from source, the server starts from the compiled index file and integrates with MCP clients through the standard mcp_settings.json configuration.

Tools provided

The server exposes a single tool to retrieve medical information.

Troubleshooting

If the server does not respond, verify that the path to the executable or built dist/index.js is correct in your mcp_settings.json. Ensure the MCP client you use supports the MCP protocol and is configured to connect to the local server.

Available tools

statpearls_disease_info

Fetches comprehensive, reliable medical information about diseases from StatPearls and returns it in Markdown format with a title, optional table of contents, and structured sections such as etiology, epidemiology, pathophysiology, clinical features, diagnosis, treatment, and prognosis.