MCP server exposing 3 tools for rijksmuseum.
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/rijksmuseum/mcp
Content-Type: application/json
{"jsonrpc":"2.0","id":1,"method":"tools/list"}rijks_search — Search the Rijksmuseum's ~800,000-object collection — Dutch Golden Age paintings, prints, drawings, photographs, Delftware, Asian art — by title, maker, object type, material, technique or object number, and get back real records: title, maker, date, materials, curatorial description, rights and a IIIF image URL. AUTHORITATIVE for what the Dutch national museum holds and how it catalogues it. Keyless. There is no free-text `q`: filters are field-scoped, and titles are catalogued in Dutch. Example: rijks_search({ creator: "Rembrandt", type: "painting", limit: 5 }).rijks_object — One Rijksmuseum object in full: title, maker, date, object type, materials, place, depicted subjects, the museum's own curatorial description, object number, rights statement and image URL. Accepts the numeric object id, its id.rijksmuseum.nl URI, or a museum object number such as "SK-C-5" (resolved for you). Keyless. Example: rijks_object({ id: "200107928" }) → The Night Watch.rijks_object_image — Image URLs for one Rijksmuseum object: the IIIF Image API 3 service, full-resolution pixel dimensions, a full-size JPEG, a scaled JPEG at a width you choose, a thumbnail, and the object's rights statement (most of the collection is public domain). Says plainly when an object has no published image rather than returning a broken link. Keyless. Example: rijks_object_image({ id: "200107928", width: 1200 }).Code samples (curl / TypeScript / one-click client install), schemas, and the live playground are on the pack page:
https://pipeworx.io/packs/rijksmuseum/
Pipeworx is an open MCP gateway connecting AI agents to live data. pipeworx.io