pubmed_search MCP server

What is pubmed_search MCP server?

pubmed_search MCP server is a server that enables searching PubMed via the Model Context Protocol (MCP). It implements resources, prompts, and tools for managing notes.

How to use pubmed_search MCP server?

Install using uv or uvx and configure in your MCP client (e.g., Claude Desktop) by adding the server definition to claude_desktop_config.json. Use the provided tools and prompts through your MCP client.

Key features of pubmed_search MCP server

• Custom note:// URI scheme for accessing individual notes
• Each note resource has name, description, and text/plain mimetype
• summarize-notes prompt with optional style argument (brief/detailed)
• add-note tool to add a new note with name and content
• Server state updates and notifies clients of resource changes

Use cases of pubmed_search MCP server

• Searching PubMed and saving relevant notes
• Summarizing a collection of notes with adjustable detail level
• Managing a personal note store via MCP resources and tools

FAQ from pubmed_search MCP server

What dependencies are required to run pubmed_search MCP server?

The server requires Python and the uv package manager. It is distributed as a Python package and uses MCP stdio transport.

How do I configure pubmed_search MCP server with Claude Desktop?

Add the server configuration to the mcpServers section in your claude_desktop_config.json. Use uv for development/unpublished servers or uvx for published servers.

Does pubmed_search MCP server support authentication or custom transports?

No authentication or custom transport is mentioned in the documentation. The server runs over stdio.

What tools and prompts does pubmed_search MCP server provide?

It provides one tool (add-note) and one prompt (summarize-notes). Resources use a custom note:// URI scheme.

Are there any known limits of pubmed_search MCP server?

The README does not specify any limits. It uses simple in-memory note storage and stdio communication.