MCP server exposing 8 tools for ncbi-eutils.
This URL is a JSON-RPC 2.0 endpoint over HTTP. Issue POST requests with a JSON-RPC body. Browsers and search crawlers land here on GET.
POST https://gateway.pipeworx.io/ncbi-eutils/mcp
Content-Type: application/json
{"jsonrpc":"2.0","id":1,"method":"tools/list"}einfo — List all available Entrez databases (omit db) or return field definitions and link names for a specific database (e.g. db=pubmed). Use before esearch to discover valid search fields and databases.esearch — Search an Entrez database (e.g. pubmed, gene, nucleotide) by query term and return a list of matching UIDs (up to 10,000) with hit count. Supports pagination via retstart and sort (e.g. pub_date).esummary — Fetch document summaries for one or more Entrez UIDs in a given database (e.g. pubmed), returning title, authors, journal, and publication date without the full record body.efetch — Full records. Returns text (XML/GenBank/FASTA/…) per rettype.elink — Links across databases.egquery — Global query across all Entrez databases.geo_series — Look up an NCBI GEO (Gene Expression Omnibus) series by its GSE accession, e.g. GSE75310 -> title, organism, number of samples and their GSM ids, platform (GPL id and name), experiment type (expression profiling by array / high-throughput sequencing, methylation, ChIP-seq...), release date, summary, PubMed ids, BioProject and SRA study links. Use for any question naming a GSE expression series.geo_search — Search NCBI GEO (Gene Expression Omnibus) for expression series (GSE) by keywords, optionally filtered by organism and experiment type. Returns the matching series with title, organism, sample count, platform, type and release date, plus the total hit count. Follow up with geo_series for one accession.Code samples (curl / TypeScript / one-click client install), schemas, and the live playground are on the pack page:
https://pipeworx.io/packs/ncbi-eutils/
Pipeworx is an open MCP gateway connecting AI agents to live data. pipeworx.io