Pipeworx pubmed

Search PubMed for biomedical articles by keyword, author, or MeSH term; retrieve article metadata and full abstracts with structured section labels.

Security Scan

The skill's requests and instructions are consistent with its stated purpose (searching PubMed), but it routes queries through an external third-party gateway (gateway.pipeworx.io) so you should evaluate that endpoint before using it with sensitive queries.

Purpose & Capability

The name/description (search PubMed, return metadata and abstracts) matches the SKILL.md. The examples and listed tools (search_pubmed, get_summary, get_abstract) all align with a PubMed search/metadata retrieval skill.

Instruction Scope

The SKILL.md directs the agent to call a third-party HTTP endpoint (https://gateway.pipeworx.io/pubmed/mcp) to perform searches and retrievals. The instructions do not ask the agent to read local files, environment variables, or unrelated system state; however, they do send user queries and results to an external host that is not the official NCBI API endpoint, which is a privacy/operational consideration.

Install Mechanism

This is an instruction-only skill with no install spec and no files to write to disk, which is the lowest-risk installation model.

Credentials

The skill declares no required environment variables or credentials. That is proportionate to the stated functionality. Note: because the skill uses a third-party gateway, it may expose query content to that remote host even though no credentials are requested.

Persistence & Privilege

The skill is not configured as always:true and requests no system-wide configuration changes. Autonomous invocation is allowed (the platform default) but is not combined with broad credentials or other elevated privileges.

Assessment

This skill appears to do what it says (search PubMed) and needs no credentials or install, but it sends queries through a third-party gateway (gateway.pipeworx.io) rather than directly to NCBI. Before installing or enabling it: 1) Verify the trustworthiness and privacy policy of gateway.pipeworx.io (who operates it, logging/retention, data sharing). 2) Avoid sending sensitive or private data in queries if you cannot trust the gateway. 3) If you prefer more control, use a skill that calls NCBI E-utilities (e.g., api.ncbi.nlm.nih.gov) directly or host a trusted proxy. 4) Do not add the gateway to any global/always-on MCP config unless you trust it. If you can provide more info about the gateway operator or hosting, I can raise or lower my confidence accordingly.

PubMed

Search and retrieve biomedical literature from the National Library of Medicine's PubMed database. Over 36 million citations from MEDLINE, life science journals, and online books.

Capabilities

Tool	Description
search_pubmed	Search by keyword, author name, or MeSH term. Returns PubMed IDs.
get_summary	Metadata for one or more articles: title, authors, journal, DOI.
get_abstract	Full abstract text with section labels (Background, Methods, Results, Conclusions).

Typical workflow

1. Search for articles on a topic with search_pubmed
2. Get metadata for the top results with get_summary
3. Pull the full abstract for the most relevant paper with get_abstract

Example: find CRISPR research

curl -X POST https://gateway.pipeworx.io/pubmed/mcp \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"search_pubmed","arguments":{"query":"CRISPR cancer therapy","limit":5}}}'

Query syntax tips

PubMed supports field-tagged searches. Some useful patterns:

• "Smith J"[Author] -- search by author
• "COVID-19"[MeSH] -- search MeSH terms
• "Nature"[Journal] -- restrict to a journal
• Combine with AND/OR: CRISPR AND (cancer OR tumor)

Add to your MCP client

{
  "mcpServers": {
    "pubmed": {
      "url": "https://gateway.pipeworx.io/pubmed/mcp"
    }
  }
}

No API key required. Powered by NCBI E-utilities.