pubmedmcp

MCP.Pizza Chef: grll

PubMed MCP is a server that integrates with the Model Context Protocol to provide seamless access to PubMed's extensive biomedical literature database. It allows users to search and fetch articles from over 35 million citations, including MEDLINE, life science journals, and online books. Leveraging the pubmedclient Python package, this MCP server facilitates real-time querying and retrieval of scientific articles, making it ideal for research, academic, and clinical applications. It can be easily configured in environments like Claude Desktop, enabling AI models to interact with up-to-date biomedical data efficiently.

Use This MCP server To

Search biomedical articles by keywords or topics Fetch detailed article metadata and abstracts Integrate PubMed data into AI research assistants Support clinical decision-making with latest literature Automate literature review for scientific research Provide real-time biomedical data to LLMs

README

PubMed MCP

A MCP server that allows to search and fetch articles from PubMed.

PubMed is a database of over 35 million citations for biomedical literature from MEDLINE, life science journals, and online books.

This MCP server relies on the pubmedclient Python package to perform the search and fetch operations.

Usage

Add the following to your claude_desktop_config.json file:

{
    "mcpServers": {
        "pubmedmcp": {
            "command": "uvx",
            "args": ["pubmedmcp@latest"],
            "env": {
                "UV_PRERELEASE": "allow",
                "UV_PYTHON": "3.12"
            }
        }
    }
}

Make sure uv is installed on your system and 'uvx' is available in your PATH (claude PATH sometimes is not the same as your system PATH). You can add a PATH key in your claude_desktop_config.json file to make sure uv is available in claude PATH.

pubmedmcp FAQ

How do I install the PubMed MCP server?

Install the server by adding it to your claude_desktop_config.json with the specified command and args, ensuring uvx and Python 3.12 are available.

What dependencies does PubMed MCP require?

It relies on the pubmedclient Python package and requires uvx to be installed and accessible in your system PATH.

Can I use PubMed MCP with different LLM providers?

Yes, it is compatible with various LLMs including OpenAI, Claude, and Gemini, as it follows the MCP standard.

How do I ensure uvx is available in my environment?

Add the uvx executable to your system PATH or specify the PATH in claude_desktop_config.json to make it accessible.

Does PubMed MCP support fetching full-text articles?

PubMed MCP primarily fetches citations and abstracts; full-text access depends on external journal subscriptions or open access availability.

Is PubMed MCP suitable for real-time querying?

Yes, it is designed to perform real-time search and fetch operations from PubMed efficiently.

How do I update PubMed MCP to the latest version?

Update by changing the version tag in the args field of your configuration, e.g., pubmedmcp@latest, and restart the server.

Can I customize search parameters in PubMed MCP?

Yes, the underlying pubmedclient package supports various search filters which can be configured through the MCP server interface.