icon for mcp server

PubMed医学文献

STDIO

用于搜索分析PubMed医学论文的服务器

Medical Literature Research Tool

An advanced Model Content Protocol (MCP) server providing tools to search, analyze, and retrieve academic medical papers from the PubMed database.

Features

  • Search for medical literature using topics and researcher names
  • Retrieve comprehensive publication details with structured metadata
  • Generate formatted citations for publications
  • Analyze researcher publication statistics and patterns
  • Advanced error handling with retry mechanisms
  • Detailed performance metrics

Installation

  1. Clone this repository:

    git clone <repository-url> cd medical-literature-tool
  2. Install dependencies:

    pip install -r requirements.txt
  3. Create a .env file in the project root if needed for configuration

Usage

  1. Start the server:

    mcp run pubmed_server.py

    For development mode:

    mcp dev pubmed_server.py
  2. Or add the server to your MCP client configuration.

API Tools

1. find_articles

Search for medical literature matching specified topics and researchers.

Parameters:

  • topics (List[str]): Medical topics or keywords to search in titles and abstracts
  • researchers (List[str]): Researcher/author names to search for
  • result_limit (int): Maximum number of results to return (default: 15)

Returns:

  • Dictionary with search results, metadata, and performance metrics

2. get_publication_details

Retrieve comprehensive details for a specific publication, including a formatted citation.

Parameters:

  • article_id (str): PubMed ID of the article to retrieve

Returns:

  • Dictionary containing detailed article metadata and citation

3. get_article_statistics

Analyze publication patterns for a specific researcher.

Parameters:

  • researcher (str): Name of the researcher/author to analyze

Returns:

  • Dictionary with publication statistics, including total count, top journals, and publication years

Technical Implementation

The server is built with a robust architecture:

  • Object-Oriented Design: Using classes for better code organization
  • Advanced Error Handling: Request retry mechanism for API reliability
  • Performance Monitoring: Timing and metrics for search operations
  • Enhanced Data Structures: Nested JSON responses with rich metadata
  • Logging System: Rotating logs with detailed error tracking
  • Modular Components: Separation of concerns between query building, API requests, and data parsing

MCP Now 重磅来袭,抢先一步体验