Met Museum MCP for AI. Search 470K+ Artworks by Keyword and Department
Gemini Works with every AI agent you already use
…and any MCP-compatible client
How this MCP server connects to your AI agent
Met Museum MCP Server provides direct access to the Metropolitan Museum of Art's open API. Your AI agent can search 470,000+ artworks using keywords, filter by department or culture, and retrieve deep metadata (dates, materials, dimensions) instantly.
It’s built for researchers who need structured data on art history without scraping websites.
What AI agents can do with Met Museum Automation
List departments
Lists all available museum departments to define your search scope.
Get object
Retrieves the full record details for a single object ID.
List objects
Returns a list of every valid, existing object ID in the collection.
Your agent runs search_objects to find multiple artworks matching criteria like artist name, culture, or material.
The agent calls list_departments to get a clean list of all 19 museum sections you can narrow your search down to.
Using get_object, the agent retrieves every piece of metadata—dimensions, dates, titles—for one specific artwork ID.
The agent executes list_objects to generate a complete list of valid object identifiers across the entire collection.
You can use get_object on any ID returned by list_objects to confirm it’s active and retrieve its current record data.
Ask an AI about this
Waiting for input…
What AI agents can do with Met Museum MCP Server: 4 Tools for Art Analysis
Use these four tools to search the collection, list departments, retrieve full records, and manage art history data programmatically.
Make your AI actually useful.
Add this MCP to Claude, Cursor, or Windsurf and your AI stops guessing. It gets real tools to look things up, take action, and handle the stuff you keep doing by hand.
Start using Met Museum on VinkiusList Departments
Lists all available museum departments to define your search scope.
Get Object
Retrieves the full record details for a single object ID.
List Objects
Returns a list of every valid, existing object ID in the collection.
Search Objects
Searches for artworks using keywords and optional filters like department or culture.
Security and governance baked right in.
Pick your AI client below to get set up. Just create a Vinkius account, subscribe, and you're instantly up and running. We handle the entire backend infrastructure, delivering out-of-the-box support for HTTPS Streamable, SSE, and OAuth2—zero messy routing required.
Claude AI
Open Claude Settings
Go to claude.ai, click your profile icon, then navigate to Customize → Connectors.
Add Custom Connector
Click the "+" button and select Add custom connector. Paste your Vinkius endpoint URL:
https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp
Replace [YOUR_TOKEN_HERE] with your token
from cloud.vinkius.com. For OAuth-protected servers, expand
Advanced settings to add credentials.
Start a conversation
Open a new chat. The Met Museum integration is available immediately — no restart needed.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with Met Museum, then connect any of our 5,100+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 5,100+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Met Museum. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Built on the Model Context Protocol (MCP) for Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This connection provides 4 powerful capabilities that interface natively with Claude, ChatGPT, Cursor, and other compatible AI platforms. No middleware. No custom integration required.
Sifting through museum data used to mean endless clicking., Solved with Vinkius AI Gateway
Before this server, getting deep metadata meant hopping between the Met website, its departmental pages, and separate research databases. You'd click into a department, then search by keyword, hit an object ID, and finally copy-paste disparate pieces of info—like dimensions from one tab and material from another—into your local spreadsheet.
Now, you just ask your agent to find it. The agent runs `search_objects`, pulling the IDs you need, and then calls `get_object` on those results in sequence. You get a clean JSON payload containing every single data point for comparison, without touching a browser.
Met Museum MCP Server: Extracting Object Data
Manual processes involve running multiple API calls (if the API was exposed that way) and then writing complex parsing scripts to stitch together the department, object ID, and metadata. This takes hours of development time just for data cleanup.
With this server, you specify the goal—like 'Find all Bronze Age figures from Egyptian Art.' The agent handles the necessary sequence: `list_departments` -> filter by Egyptian Art ID -> run `search_objects` with 'figure' -> finally pull full records using `get_object`. It’s a single, reliable workflow.
What your AI can actually do with this
You're running an art history project, and you need structured data from the Met Collection—not just pretty pictures. This server gives your AI agent direct access to over 470,000 artworks and their metadata. You get clean JSON records every time; no scraping websites necessary.
To start scoping your research, you first call list_departments. This tool spits out a complete list of all nineteen museum departments, letting you narrow down the scope of what you're looking for right off the bat. Once you know your department, you use search_objects to run targeted searches. You can hit it with keywords—like an artist's name or a specific material—or filter by culture and department to find multiple artworks that fit your criteria.
When you pull up a list of potential objects from search_objects, you need the deep background info. You use get_object on any single object ID to retrieve every piece of metadata available for that artwork, including precise dimensions, creation dates, and titles. This gives you the full record details for one specific item.
If you're building a database or checking inventory across the whole collection, you execute list_objects. That tool returns a massive list of valid object identifiers across all 470,000+ pieces. You can then take any ID returned by list_objects and run it through get_object. This confirms that the object ID is active in the system and grabs its current record data for validation.
Every time you check an ID this way, you're making sure your dataset stays accurate.
The entire process flows like this: You use list_departments to define your search area; you run search_objects with keywords or filters (like culture or material) to find a list of relevant object IDs; and then, for every single one that matters, you call get_object to pull the full metadata—dimensions, dates, titles—or use list_objects followed by get_object if you just need to confirm the ID's existence.
You never have to worry about complex API calls or parsing messy web pages; your agent handles it all with structured JSON output.
019e5d35-0403-719a-96ff-d9f4e28d5843 
Here's how it actually works
The bottom line is that your AI client chains these tools together: scope it down, search within the scope, and then extract the deep data you need.
First, your agent runs list_departments to get the scope. This defines which museum wings (e.g., Egyptian Art) you want to focus on.
Next, the agent uses search_objects, passing in a keyword and a department ID found in step one. This filters down the 470k collection into manageable results.
Finally, if you need full details on one result, the agent calls get_object with the specific object ID to pull all metadata.
Who is this actually for?
This server is for academic researchers, digital humanities students, or creative professionals who deal with massive, structured datasets. If your job involves cross-referencing historical objects with modern content—or if you're tired of manually browsing museum websites—you need this.
Uses search_objects to find all examples of a specific technique (e.g., 'sfumato') across different cultures and then uses get_object to compare their exact dimensions.
Calls list_departments first, then structures the data pipeline using search_objects to pull public domain imagery for inspiration on a specific theme or era.
Runs list_object and iterates through IDs, analyzing metadata trends (material, date range) across different geographic regions.
What Changes When You Connect
Get structured, deep metadata instantly. Instead of guessing what details a piece has, get_object pulls titles, materials, dimensions, and dates for any given ID.
Define your search scope immediately. Using list_departments gives you the 19 main museum sections right away, letting you filter results before running a single keyword search.
Avoid massive list dumps. While list_objects shows every valid ID, always pair it with search_objects and a department ID to keep your queries focused and efficient.
Streamline research for creatives. You can use search_objects combined with image data to find public domain art inspiration based on specific historical periods or mediums.
Build complex workflows easily. The server allows chaining calls: list departments -> search objects by department ID -> get object details, all in one agent run.
See it in action
Comparing Period Styles
An art historian needs to compare three specific sculptures. They first call list_departments to confirm the 'Greek and Roman' ID. Then, they use search_objects with keywords like 'bust' and that department ID. Finally, they run get_object on each of the resulting IDs to pull identical metadata fields for a direct comparison.
Curating a Design Moodboard
A designer is working on an industrial-era piece and needs inspiration from public domain art. They use search_objects with 'metal' or 'industrial' as keywords, limiting the search to specific departments they know contain that material. This provides high-quality source images without needing a visual database.
Analyzing Collection Growth
A data scientist wants to see how many objects were added last year and what their dominant materials are. They run list_objects to get the full list, then filter that dataset using metadata fields (date of update, material) to analyze collection trends.
Validating Object IDs
Before building a massive research database, you need to know which object IDs are actually active. You call list_objects once, and then use the resulting list to confirm every ID works by running get_object on a sample batch.
The honest tradeoffs
Over-relying on keywords alone
Just asking the agent to 'Find all statues from Greece.' This risks returning millions of irrelevant results across unrelated departments, bogging down the call and losing focus.
Always start by running list_departments to find the specific department ID (e.g., Greek Art). Then, run search_objects, passing both your keyword ('statue') AND that confirmed department ID.
Treating object IDs as simple keys
Assuming every random number you find is a valid art object. Many numbers are invalid or point to empty records, wasting calls.
First, use list_objects to get the list of valid IDs. Then, test them with get_object before building any permanent workflow around them.
Forgetting metadata depth
Getting a search result and only seeing the title and artist name. This leaves you unable to compare dimensions or dates needed for academic work.
After search_objects gives you IDs, always follow up with get_object on the key results. That single call pulls all available data: materials, dimensions, titles, etc.
When It Fits, When It Doesn't
Use this server if your task involves structured comparison or large-scale academic querying of historical artifacts. The workflow must be precise: you need to narrow scope first—use list_departments to find the container (the department) and then use that ID with search_objects. Don't use it if you just want a general overview; for that, a web search is faster. Also, don't rely solely on keywords; always cross-reference your query using one of the department IDs found via list_departments. If you only have an object ID and need all its info, skip straight to get_object. Never assume any single tool covers everything—it's a multi-step process.
Questions you might have
How do I start searching for objects using the Met Museum MCP Server? +
You should first call list_departments to get a list of valid departments. Then, use that department ID in your search_objects query along with keywords.
What is the difference between `get_object` and `search_objects`? +
search_objects finds multiple potential objects based on criteria. get_object requires a specific ID and pulls every single detail for only that one object.
Can I use Met Museum MCP Server to find images of artworks? +
Yes, the API provides high-resolution URLs for objects confirmed as being in the public domain. This data is included when you run get_object or search_objects.
Do I need to know the object ID before using Met Museum MCP Server? +
No. If you don't know the ID, start with a broader search by running list_departments, then use that department in search_objects.
How do I filter my search results using the `list_departments` tool? +
The list_departments tool returns all 19 valid museum departments and their corresponding IDs. You use those returned department IDs as filters when calling search_objects or get_object, which helps narrow down millions of records to a specific collection area.
If I run many searches, how should I manage my usage with the Met Museum MCP Server? +
The server observes standard API rate limits. If your agent makes too many requests in a short time, you'll get an error. To avoid this, build batch calls where possible or add a slight delay between successive operations.
What is the purpose of using `list_objects`? +
The list_objects tool provides a current list of valid Object IDs across the entire collection. This function lets you audit your data and confirm which specific object identifiers are active in the Met's digital archive.
Do I need to pass an API key when calling `get_object`? +
You must provide an identifier as required by your proxy setup. If the museum's public access mandates no specific key, simply use 'PUBLIC' as the identifier string in your agent’s workflow.
How can I search for artworks by a specific artist that have images available? +
You can use the search_objects tool. Set the q parameter to the artist's name and set hasImages to true. This will return a list of Object IDs matching those criteria.
Can I get the full historical record and image URL for a specific piece of art? +
Yes! Use the get_object tool with the specific objectID. It returns detailed metadata including the artist, medium, dimensions, and primaryImage or additionalImages URLs.
How do I find out which departments exist in the Met Museum? +
Use the list_departments tool. It will provide a complete list of all valid departments along with their unique departmentId, which you can then use to filter your searches.
We've already built the connector for Met Museum. Just plug in your AI agents and start using Vinkius.
No hosting. No infrastructure. No complex setup.
All 4 tools are live and waiting.
You're up and running in seconds.
Gemini Vinkius gives your AI agents access to the full catalog of app connectors, all fully managed, secure, and enterprise-ready. One subscription, every tool you need.
Built, hosted, and secured by Vinkius. You just connect and go.