PubMed MCP
README.md
# 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](https://github.com/grll/pubmedclient) Python package to perform the search and fetch operations.
## Credits
This project is maintained by **bachstudio**.
Original project: This is based on the [pubmedmcp](https://github.com/grll/pubmedmcp) project originally developed by [grll](https://github.com/grll).
## Usage
Add the following to your `claude_desktop_config.json` file:
```json
{
"mcpServers": {
"pubmedmcp": {
"command": "uvx",
"args": ["bachstudio-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.TDQS
A3.6/5.0
Scored across 1 tool
Disambiguation5/5
With only one tool, there is no possibility of confusion between tools. The single tool is inherently distinct.
Naming Consistency5/5
There is only one tool, so naming consistency is not a concern. The name 'search_abstracts' follows a clear verb_noun pattern.
Tool Count2/5
A single tool for a comprehensive biomedical database like PubMed feels significantly under-scoped. Users would benefit from additional tools for specific operations (e.g., fetching full details, filtering by fields).
Completeness2/5
The server lacks tools for specialized queries, structured data retrieval, or pagination. While the single tool returns rich data, it does not cover common PubMed operations like fetching a specific article by ID or searching by author/journal.
Maintenance
ActivityInactive
ResponsivenessNo issues