Pubmed MCP Server

CLI and Python SDK for biomedical literature search with MeSH, PICO, and intelligent query expansion

Local serverstdioPython

What is the Pubmed MCP server?

Connect Pubmed to Claude, Cursor or any other MCP client and it stops being a tab you switch to. CLI and Python SDK for biomedical literature search with MeSH, PICO, and intelligent query expansion. The pubmed mcp server is what makes that connection.

What the server does

CLI and Python SDK for biomedical literature search, query analysis, citation export, full-text lookup, and multi-source academic enrichment.

Available tools

The toolset is worth reading before you wire it up, because it tells you what the integration is really for:

  • search — Search biomedical literature across PubMed and optional academic sources
  • strategy — Generate MeSH expansion and search strategy materials
  • analyze-query — Analyze intent, complexity, identifiers, PICO, and source recommendations without running a search
  • details — Fetch detailed metadata for one or more PMIDs
  • export — Export citations as BibTeX, RIS, CSV, MEDLINE, CSL JSON, or JSON
  • fulltext — Locate full-text access through available source resolvers
  • text-mined — Retrieve text-mined biomedical terms when source data is available
  • related — Find related articles for a PMID
  • metrics — Fetch citation metrics where supported
  • timeline — Build research timeline summaries from a query or PMID set
  • cite-tree — Explore citation/reference relationships around a root PMID
  • pico — Parse a clinical question into PICO-style components

Installation

pubmed-search on PyPI is all you need. Most clients run it directly, so configuration is a few lines and a restart.

Credentials and setup notes

Configuration is passed through the environment: NCBI_EMAIL, NCBI_API_KEY. Treat anything key-shaped as a real credential — scope it to the minimum the server needs, and rotate it if it ever lands in a shared config.

Where it fits

Among the developer tooling options, the useful question is rarely "what can it do" but "what does it cost you to run" — permissions, credentials, and how much of your context its toolset consumes. Pubmed's toolset — search, strategy, analyze-query and 9 more — is a fair guide to whether it matches your workflow.

SyncDev reviews every entry in this directory against the project's own documentation before publishing, and revisits them as servers change.

Worth knowing first

  • It runs with your machine's permissions. That is convenient and also the reason to think about what you point it at before you approve a tool call.
  • With 12 tools registered it takes up a noticeable share of the context window; turn it off in projects that never touch Pubmed.
  • Missing credentials fail quietly in some clients — if no tools show up, check the environment block first.
  • Keep per-call confirmation enabled while you learn its behaviour; it is the cheapest safeguard you have.

Available tools

ToolWhat it does
searchSearch biomedical literature across PubMed and optional academic sources.
strategyGenerate MeSH expansion and search strategy materials.
analyze-queryAnalyze intent, complexity, identifiers, PICO, and source recommendations without running a search.
detailsFetch detailed metadata for one or more PMIDs.
exportExport citations as BibTeX, RIS, CSV, MEDLINE, CSL JSON, or JSON.
fulltextLocate full-text access through available source resolvers.
text-minedRetrieve text-mined biomedical terms when source data is available.
relatedFind related articles for a PMID.
metricsFetch citation metrics where supported.
timelineBuild research timeline summaries from a query or PMID set.
cite-treeExplore citation/reference relationships around a root PMID.
picoParse a clinical question into PICO-style components.

How to install the Pubmed MCP server

{
  "mcpServers": {
    "pubmed-search-1": {
      "command": "uvx",
      "args": ["pubmed-search"],
      "env": {
        "NCBI_EMAIL": "your-value",
        "NCBI_API_KEY": "your-value"
      }
    }
  }
}

Add to claude_desktop_config.json, then restart Claude Desktop.

Configuration

VariableDescriptionRequired
NCBI_EMAILConfiguration value read at startup.Optional
NCBI_API_KEYCredential the server authenticates with.Yes

Example prompts to try

  • Use Pubmed to search.
  • Use Pubmed to strategy.
  • Use Pubmed to analyze-query.

Frequently asked questions

It connects Pubmed to MCP-compatible AI assistants such as Claude and Cursor, exposing 12 tools (search, strategy, analyze-query, and more) that the assistant can call on your behalf. Instead of copying data back and forth by hand, the assistant works with Pubmed directly.