MCP server exposing 4 tools for ema-article57.
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/ema-article57/mcp
Content-Type: application/json
{"jsonrpc":"2.0","id":1,"method":"tools/list"}ema_article57_search — Search the EMA Article 57 database — every medicine with a valid marketing authorisation anywhere in the EEA (centrally, mutually-recognised, decentralised AND purely nationally authorised), one row per product x authorisation-country. Answers "which EEA countries is <product> authorised in", "who is the marketing-authorisation holder / pharmacovigilance system location for <product> in <country>", "which products does <company> hold in the EEA". Free-text query matches product name, active substance and holder name. Real offset+limit pagination with a total row count in the response — this is a 164k-row register, not a top-N sample.ema_article57_product — The full Article 57 record(s) for one product by name — every EEA country it is authorised in, with the marketing-authorisation holder and pharmacovigilance system location per country. Article 57 carries no indication text and no per-product URL, so this returns the authoritative EMA source document URL rather than a paraphrased indication.ema_article57_recent_changes — Products added to or removed from the Article 57 register since a given date, computed by diffing our dated snapshots of the EMA export (the source itself carries no change-date column, so there is no other way to answer "what changed"). Returns comparable:false with an explicit reason, never a guessed empty diff, when no snapshot exists at or before the requested date.ema_article57_dataset_info — Dataset metadata: the authoritative EMA bulk-download URL, current row count, this pack's snapshot dates on record, and the full list of distinct authorisation-country values used by the country filter. Use this before ema_article57_search when you need the exact country spelling or want the bulk file instead of paginated rows.Code samples (curl / TypeScript / one-click client install), schemas, and the live playground are on the pack page:
https://pipeworx.io/packs/ema-article57/
Pipeworx is an open MCP gateway connecting AI agents to live data. pipeworx.io