Flaex Logo
Flaex.ai
Back to MCP Directory
Simple Pubmed
Model Context Protocol low risk

Simple Pubmed

Provides access to PubMed articles and abstracts via the Entrez API, enabling literature searches and full-text retrieval for AI agents.

Connections & Capabilities

Connects To

GitHubTwitter

Capabilities

read

Quickstart

Install

npx -y @smithery/cli install mcp-simple-pubmed --client claude

Config

{
  "mcpServers": {
    "simple-pubmed": {
      "command": "python",
      "args": ["-m", "mcp_simple_pubmed"],
      "env": {
        "PUBMED_EMAIL": "your-email@example.com",
        "PUBMED_API_KEY": "your-api-key" 
      }
    }
  }
}

Exposed MCP Tools (6)

safe
Search PubMed

Searches the PubMed database using keywords, MeSH terms, author names, date ranges, and Boolean operators.

Read-only search operation.

safe
Get Full Text

Downloads the full text of an article from PubMed Central, when available.

Read-only retrieval of publicly available data.

safe
Access Abstracts

Retrieves article abstracts and metadata via resource URIs.

Read-only access to article metadata.

safe
Systematic Review Search

Generates comprehensive search strategies for systematic reviews using MeSH terms and synonyms.

Generates search queries; no data modification.

safe
PICO Search

Builds clinical question searches using the PICO framework.

Generates search queries; no data modification.

safe
Author Search

Finds publications by a specific author with proper name formatting.

Generates search queries; no data modification.

Safety Assessment

This server is relatively safe due to its read-only nature and reliance on a well-established API. Risks are primarily related to credential management (email, API key) and the potential for AI agents to misinterpret retrieved data. It is safe to use for literature searches but requires careful handling of API keys.

  • Read-only access to PubMed data
  • No direct modification of external systems
  • Relies on NCBI's API for data retrieval
  • No local data storage or processing of sensitive information
  • Rate limiting is managed (though not enforced in the code)
  • Requires an email address for NCBI API usage
  • Optional API key adds a potential credential exposure risk
  • Full-text retrieval returns XML, which might be unexpected
  • Potential for misinterpretation of search results by AI agents
  • Dependence on external NCBI service availability