home / mcp / pubmed mcp server
Provides search and fetch capabilities for PubMed articles via an MCP server.
Configuration
View docs{
"mcpServers": {
"grll-pubmedmcp": {
"command": "uvx",
"args": [
"pubmedmcp@latest"
],
"env": {
"UV_PYTHON": "3.12",
"UV_PRERELEASE": "allow"
}
}
}
}PubMed MCP is an MCP server that lets you search PubMed and fetch articles programmatically. It uses the pubmedclient Python package under the hood, enabling you to query a vast biomedical literature database and retrieve article details directly through your MCP client.
To use PubMed MCP, start the local MCP server and connect to it with your MCP client. The server is configured to run via the uvx runtime, and it exposes endpoints that let you perform PubMed searches and fetch article data. Ensure the uv runtime is available on your system and that the server command is in your PATH. Once the server is running, you can issue search queries and fetch results from PubMed through your MCP client just like you would with any other MCP data source.
Prerequisites and initial setup for PubMed MCP are straightforward if you follow these steps.
1) Ensure the UV runtime is installed on your system and that the uvx command is accessible in your PATH. If your environment uses a dedicated Claude PATH, you may need to add a PATH entry in your configuration to make uvx available to Claude.
2) Add the MCP server configuration to your Claude desktop config. Use the following snippet exactly as shown to enable PubMed MCP.
{
"mcpServers": {
"pubmedmcp": {
"command": "uvx",
"args": ["pubmedmcp@latest"],
"env": {
"UV_PRERELEASE": "allow",
"UV_PYTHON": "3.12"
}
}
}
}The required environment variables shown in the server configuration are UV_PRERELEASE and UV_PYTHON. You can adjust their values as needed, but the example uses UV_PRERELEASE set to allow and UV_PYTHON set to 3.12.
- The PubMed MCP server relies on the PubMed data source, which includes millions of citations and related biomedical literature.
Perform PubMed searches using queries and filters to find relevant articles.
Retrieve detailed article data for specific PubMed entries identified in a search.