Pubmed MCP Server

🚀 PubMed Analysis MCP Server

This is a newly developed MCP server tailored for medical students. Its features are still under improvement. We warmly welcome your suggestions and contributions!

A professional MCP server for analyzing PubMed medical literature, which enables researchers to quickly grasp the dynamics of medical research.

🚀 Quick Start

1. Install dependencies: Run pip install -r requirements.txt or use the uv environment for installation.
2. Configure the mcp.json file and add the PubMed server configuration.
3. Start the MCP server and initiate analysis tasks.

✨ Features

• Literature Retrieval: Supports PubMed advanced search syntax and allows setting date ranges and the number of results.
• Hotspot Analysis: Counts keyword frequencies, identifies popular research directions, and aggregates relevant literature.
• Trend Tracking: Tracks the frequency changes of keywords over time to reveal the evolution of research trends.
• Publication Count: Provides flexible time - period settings to analyze changes in the number of publications.
• Comprehensive Reports: Generates analysis reports containing hotspots, trends, and statistics with a single click.

📦 MCP Tools

1. search_pubmed

Search PubMed and save the results.

Key parameters:

• email: Your email (required)
• advanced_search: PubMed search query (required, same as advanced search syntax)
• max_results: Maximum number of results (default: 1000)

2. list_result_files

List available result files.

3. analyze_research_hotspots

Analyze research hotspots.

Key parameters:

• filename: Result filename (required)
• top_n: Number of keywords to analyze (default: 20)

4. analyze_research_trends

Analyze research trends.

5. analyze_publication_count

Analyze the number of publications.

6. generate_comprehensive_analysis

Generate a comprehensive analysis report.

💻 Usage Examples

Basic Usage

# Install dependencies
pip install -r requirements.txt
# Or use uv
uv pip install -r requirements.txt

Advanced Usage

Write mcp.json

Since I'm used to the uv virtual environment, I directly run Python files using the Python path here.

// Add the following configuration in mcp.json (for Windows)
"PubMed": {
        "command": "cmd",
        "args": [
          "/c",
          "path/to/python.exe",
          "path/to/server.py"
        ]
    }
// For example, my mcp.json file looks like this
{
    "mcpServers": {
      "fetch":{
        "command": "cmd",
        "args": [
          "/c",
          "uvx",
          "mcp-server-fetch"
        ]
      },
      "PubMed": {
        "command": "cmd",
        "args": [
          "/c",
          "path/to/python.exe",
          "path/to/server.py"
        ]
      }
    }
  }

📚 Documentation

Notes

• Project Description: This project is a newly developed MCP server designed to assist medical researchers in analyzing PubMed literature.
• Result Saving: Analysis results will be saved in the specified folder. Please ensure the path is correct.
• Dependency Installation: Ensure a stable network connection when installing dependencies and choose the appropriate Python version as needed.

Thank you for your support and understanding!