PubMed Search MCP server

The MCP server for PubMed searching enables interactions with PubMed through the Model Context Protocol (MCP), allowing for seamless integration with AI assistants like Claude. It provides a structured way to query medical literature through a standardized interface.

Installation

For Claude Desktop Users

To integrate the PubMed search MCP server with Claude Desktop, you'll need to modify your configuration file based on your operating system:

• macOS: ~/Library/Application\ Support/Claude/claude_desktop_config.json
• Windows: %APPDATA%/Claude/claude_desktop_config.json

Configuration Options

For Published Version

If you're using the published version of the server:

"mcpServers": {
  "pubmed_search": {
    "command": "uvx",
    "args": [
      "pubmed_search"
    ]
  }
}

For Development Version

If you're working with an unpublished/development version:

"mcpServers": {
  "pubmed_search": {
    "command": "uv",
    "args": [
      "--directory",
      "/Users/nofuture/Documents/GitHub/pubmed_search",
      "run",
      "pubmed_search"
    ]
  }
}

Make sure to update the directory path to match your local setup.

Using the Server

Available Functionality

The PubMed search MCP server provides the following capabilities:

Note Storage System

• Uses a custom note:// URI scheme for accessing individual notes
• Notes include name, description, and text content
• All notes use the text/plain MIME type

Available Prompt

The server includes one prompt:

• summarize-notes: Creates summaries of all stored notes
  • Accepts an optional "style" argument (brief/detailed) to control detail level
  • Combines all current notes with your style preference into a single prompt

Available Tool

The server provides one tool:

• add-note: Adds a new note to the server
  • Required arguments:
    • "name": The title for your note
    • "content": The text content of your note
  • The server updates its state and notifies clients of resource changes

Working with Notes

1. Adding Notes: Use the add-note tool to store information from your PubMed searches
2. Summarizing: Use the summarize-notes prompt to get a consolidated view of your stored notes
3. Accessing Individual Notes: Reference specific notes using the note:// URI scheme

Best Practices

• Give your notes descriptive names for easy identification
• Use the style parameter with summarize-notes to control the level of detail in summaries
• Consider organizing related information across multiple notes for better structure

Troubleshooting

If you encounter issues with the server:

• Verify your configuration file has the correct format and paths
• Ensure you have the necessary permissions to access the specified directories
• Check that you have the correct version of the package installed

How to add this MCP server to Cursor

There are two ways to add an MCP server to Cursor. The most common way is to add the server globally in the ~/.cursor/mcp.json file so that it is available in all of your projects.

If you only need the server in a single project, you can add it to the project instead by creating or adding it to the .cursor/mcp.json file.

Adding an MCP server to Cursor globally

To add a global MCP server go to Cursor Settings > MCP and click "Add new global MCP server".

When you click that button the ~/.cursor/mcp.json file will be opened and you can add your server like this:

{
    "mcpServers": {
        "cursor-rules-mcp": {
            "command": "npx",
            "args": [
                "-y",
                "cursor-rules-mcp"
            ]
        }
    }
}

Adding an MCP server to a project

To add an MCP server to a project you can create a new .cursor/mcp.json file or add it to the existing one. This will look exactly the same as the global MCP server example above.

How to use the MCP server

Once the server is installed, you might need to head back to Settings > MCP and click the refresh button.

The Cursor agent will then be able to see the available tools the added MCP server has available and will call them when it needs to.

You can also explictly ask the agent to use the tool by mentioning the tool name and describing what the function does.