- DATA
- RECORDS
- HUMAN
- LOCAL
Short answer
How can I pull Rijksmuseum metadata into my AI?
Your AI uses LDES streams and OAI-PMH requests to grab structured records directly from the museum's repositories. You'll get clean, machine-readable metadata ready for your analysis.
Data outcomes
Where your data lands.
The AI moves data from museum streams into your specific workflows.
DATA
Structured datasets
The AI pulls specific collections or datasets. You get organized metadata that fits your existing schema.
RECORDS
OAI-PMH records
The AI requests individual or batch records via standard protocols. This brings raw museum data into your chat.
HUMAN
Manual curation
The AI identifies interesting metadata patterns. You decide which specific pieces of heritage data are worth keeping.
LOCAL
Your own files
Once the AI has the data, you can move it into your own databases or documents. The AI acts as the bridge.
The workflow
What your AI does when the request arrives.
The AI acts as a bridge between museum protocols and your chat interface.
Identify collections
The AI looks through available LDES collections to find the specific heritage sets you need.
get_ldes_collectionFetch specific datasets
It targets a specific LDES dataset to pull the structured metadata you're looking for.
get_ldes_datasetRequest legacy records
For older or different repository formats, the AI uses standard requests to pull metadata.
oai_pmh_requestFormat for use
The AI cleans the incoming metadata so it's actually useful for your specific questions.
get_ldes_dataset
Try it
Copy these to start.
Use these prompts to kick off the data collection.
Starting points
Swap out the collection names or dataset IDs for the specific Rijksmuseum sets you want to study.
Rijksmuseum Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_6YIkIaAhivuXu3HoHaQeJy3j71TH7KdxS3XNAZ2V/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — Rijksmuseum capabilities are ready to use.
{
"mcpServers": {
"rijksmuseum-mcp": {
"url": "https://edge.vinkius.com/vk_preview_6YIkIaAhivuXu3HoHaQeJy3j71TH7KdxS3XNAZ2V/mcp"
}
}
}List the available LDES collections from the Rijksmuseum.
Get the metadata for the specific LDES dataset named 'master_collection'.
Run an OAI-PMH request to fetch records from the museum repository.
Summarize the metadata found in the latest LDES dataset collection.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect Rijksmuseum once, then ask.
Just link the Connector to your client. Your credentials stay encrypted, and you can start asking for data immediately.
Connect Rijksmuseum to your AIFAQ
How this task behaves.
- 01
Can the AI download high-resolution images?
No. The AI only pulls structured metadata through LDES and OAI-PMH. It doesn't handle raw image files.
- 02
Can the AI change metadata in the Rijksmuseum database?
No. The AI is read-only. It can only fetch data, not modify the museum's original records.
- 03
Does the AI handle the OAI-PMH handshake?
Yes. The AI manages the request and the response parsing automatically.
- 04
Can I use this to update my own local database?
The AI can fetch the data and format it for you, but you are responsible for saving that data into your own systems.
- 05
Will the AI decide which metadata is important?
The AI can suggest patterns, but you have to be the one to decide which data is actually useful for your work.
More questions about Rijksmuseum? The Connector page answers them. See everything the Rijksmuseum Connector can do
Connect Rijksmuseum to Claude, Cursor, ChatGPT & more
