pubmed-mcp-server and pubmed-search-mcp

pubmed-mcp-server
58
Established
pubmed-search-mcp
52
Established
Maintenance 13/25
Adoption 8/25
Maturity 18/25
Community 19/25
Maintenance 10/25
Adoption 12/25
Maturity 18/25
Community 12/25
Stars: 65
Forks: 21
Downloads:
Commits (30d): 0
Language: TypeScript
License: Apache-2.0
Stars: 5
Forks: 1
Downloads: 1,860
Commits (30d): 0
Language: Python
License:
No risk flags
No risk flags

About pubmed-mcp-server

cyanheads/pubmed-mcp-server

MCP server for the NCBI E-utilities API. Search PubMed, fetch article metadata, generate citations, explore MeSH terms, and discover related research. Runs over stdio or HTTP. Deployable to Cloudflare Workers.

Implements a complete NCBI E-utilities wrapper with seven specialized tools for article search, metadata fetching, full-text retrieval, and citation generation, built on the `@cyanheads/mcp-ts-core` framework. Features NCBI-compliant sequential request queuing, a custom XML parser handling PubMed's inconsistent structure, and hand-rolled citation formatters (APA, MLA, BibTeX, RIS) with zero external dependencies for Workers compatibility. Designed as an MCP client integration—deployable via stdio, HTTP, or Cloudflare Workers—with optional OpenTelemetry tracing and pluggable authentication.

About pubmed-search-mcp

u9401066/pubmed-search-mcp

🔬 Professional MCP server for biomedical literature research — 40 tools, multi-source search (PubMed, Europe PMC, CORE, OpenAlex), full-text access, citation networks, PICO analysis, and more

Employs a Domain-Driven Design architecture with a vocabulary translation layer that automatically converts between natural language, MeSH, ICD-10, and database-specific terminologies, enabling unified querying across fragmented APIs. Integrates directly with NCBI, Europe PMC, CORE, and OpenAlex via stdio transport with automatic credential handling, while providing research timeline detection, landmark paper identification, and citation network analysis for comprehensive literature mapping.

Scores updated daily from GitHub, PyPI, and npm data. How scores work