
PubMed MCP Server
Access and search open access PubMed research articles with detailed abstracts, free full-text links, and built-in rate limiting for efficient Model Context Protocol (MCP) integration.
Supercharge Your AI with PubMed MCP Server
Unlock the full potential of PubMed MCP Server through LangDB's AI Gateway. Get enterprise-grade security, analytics, and seamless integration with zero configuration.
Free tier available • No credit card required
PubMed MCP Server
An MCP server implementation for accessing PubMed data with focus on open access content.
Features
- Search PubMed database for research articles
- Filter for open access content
- Get detailed article information including abstracts
- Find free full-text links where available
- Built-in rate limiting for API compliance
Installation
npm install @rikachu225/pubmed-server
Usage
import PubMedServer from '@rikachu225/pubmed-server'; // Search for recent open access papers const results = await PubMedServer.getLatestOpenAccess({ topic: 'machine learning', days: 30, maxResults: 10 }); // Basic search with options const searchResults = await PubMedServer.search({ query: 'cancer treatment', maxResults: 20, sort: 'relevance', filterOpenAccess: true });
API Reference
search(options)
Search for articles with specified criteria
query
: Search termsmaxResults
: Maximum number of results (default: 10)sort
: Sort order ('relevance' or 'date')filterOpenAccess
: Whether to filter for open access content (default: true)
getLatestOpenAccess(options)
Get recent open access papers
topic
: Search topicdays
: Number of days to look back (default: 30)maxResults
: Maximum number of results (default: 10)
getOpenAccessLinks(pmid)
Get available free full-text links for an article
pmid
: PubMed ID of the article
Rate Limiting
The server implements automatic rate limiting to comply with NCBI's guidelines (maximum 3 requests per second).
License
MIT
mcp-pubmed-server
Discover shared experiences
Shared threads will appear here, showcasing real-world applications and insights from the community. Check back soon for updates!