Smithsonian Open Access MCP for AI. Access millions of global museum and scientific records.

Q: How do I find a record across all Smithsonian units using searchrecords?

You simply instruct your agent to run searchrecords with your general query. This tool queries the entire EDAN network, giving you results from every department.

Q: Is there a way to filter my search by art or history using searchcategory?

Yes, use searchcategory. You specify the desired unit (like 'Art') and the tool filters all subsequent searches to only show records from that specific taxonomy.

Q: What information does getcontent actually retrieve for an ID?

getcontent pulls comprehensive, deep metadata. It includes provenance, detailed descriptive text, and links to the digital assets associated with that record.

Q: Does searchcategory allow filtering by non-museum data types, like scientific measurements?

No, searchcategory maps only to defined Smithsonian collections and units. It restricts search results strictly to established cultural or historical metadata fields for accuracy.

Claude

ChatGPT

Cursor

Gemini

Windsurf

VS Code

JetBrains

Vercel

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

Connect to your AI in seconds.

Smithsonian Open Access connects your AI client to millions of museum records and scientific artifacts from the Smithsonian Institution's network.

You query historical data using `search_records` for broad results, filter down with `search_category`, and pull full metadata payloads with `get_content`.

It's programmatic access to global cultural heritage.

What your AI can do

Search category

Narrows down searches by limiting results to predefined units or fields like 'Art' or 'Science'.

Get content

Pulls all detailed metadata (provenance, description) for a specific object ID in the collection.

Search records

Performs a broad search query for records across all Smithsonian departments.

Global Record Search

Find museum records across all Smithsonian departments using search_records with general keywords or concepts.

Taxonomy Filtering

Limit your search scope to specific fields, like 'Art' or 'History,' by invoking the search_category tool.

Metadata Extraction

Pull all detailed information—provenance, descriptions, and asset links—for a single object using its unique ID with get_content.

Ask an AI about this

Included with Plan

Waiting for input…

AI Agent

Smithsonian Open Access MCP Server: 3 Tools for Research

These tools allow your AI client to perform complex data retrieval tasks by searching broad records, filtering specific categories, or pulling deep metadata payloads.

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 Smithsonian Open Access on Vinkius

Search Category

Narrows down searches by limiting results to predefined units or fields like 'Art' or 'Science'.

Get Content

Pulls all detailed metadata (provenance, description) for a specific object ID in...

Search Records

Performs a broad search query for records across all Smithsonian departments.

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 Smithsonian Open Access integration is available immediately — no restart needed.

Antigravity

Configure Agent Environment

Open your Antigravity agent's workspace configuration or mcp-servers.json file.

Bind the Endpoint

Add the Vinkius endpoint URL to your agent's MCP connections list:

"mcp_servers": {
  "smithsonian-open-access": {
    "serverUrl": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
  }
}

Provide your secure token in place of [YOUR_TOKEN_HERE] to ensure your agent requests are authenticated.

Execute

Start your Antigravity session. The agent will autonomously discover and utilize the Smithsonian Open Access tools with full Vinkius guardrails applied.

VS Code Copilot

⚡

One-Click Install (Recommended)

In your Vinkius Dashboard, simply click the Add to VS Code button for this server. We'll automatically configure your local workspace.

Or configure manually

Open MCP Settings

Open VS Code, press Ctrl/Cmd + Shift + P, and search for GitHub Copilot: MCP Servers.

Add Server Config

Add the Vinkius endpoint configuration to your mcp-servers.json file:

"smithsonian-open-access": {
  "url": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
}

Ensure you replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com.

LangChain

Install Dependencies

Install the LangChain MCP adapters for your environment:

pip install langchain-mcp-adapters

Connect the Server

Use the SSEClient in LangChain to connect to the Vinkius managed endpoint:

from langchain_mcp_adapters.client import SSEClient

# Connect to Vinkius
client = SSEClient(url="https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp")
tools = client.get_tools()

CrewAI

Define the Tool

Load the Vinkius MCP tools into your CrewAI agents:

from crewai import Agent
from mcp_crewai import MCPTool

# Connect securely to Vinkius
vinkius_tools = MCPTool(url="https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp")

# Assign to Agent
researcher = Agent(
    role='Data Researcher',
    tools=vinkius_tools.get_all()
)

Execute Task

Run your CrewAI process. The agent will autonomously route tasks to the Vinkius managed server.

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

Start building

Make Your AI Do More

Start with Smithsonian Open Access, 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

Smithsonian Open Access MCP server cover

Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Smithsonian Open Access. 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

Your data is protected. See how we built it.

Works with 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 3 powerful capabilities that interface natively with Claude, ChatGPT, Cursor, and other compatible AI platforms. No middleware. No custom integration required.

Manually researching global archives is a nightmare of tabs and redirects.

Right now, finding solid primary source material means navigating institutional websites. You search on one page, get 20 results, copy the ID, switch to another tab, and then use a separate form just to pull the full metadata payload. It's slow, and you constantly risk losing track of your original query context.

With this MCP server, you ask for what you need in plain language. The agent handles the hand-off: it runs `search_records` to find candidates, filters them with `search_category`, and then automatically uses `get_content` on promising IDs. You get a clean data object, not 10 browser tabs.

Smithsonian Open Access MCP Server: Get full artifact metadata.

Before this server, if you needed the detailed provenance and asset links for an item ID, you'd have to navigate through multiple departmental forms or guess which database held the information. It was a manual guessing game that often resulted in incomplete data sets.

Now, when your agent pulls an ID from any search tool, it can pass that single identifier directly into `get_content`. This gives you the complete metadata package—the full story—without ever touching another website.

Support 24/7 support@vinkius.com ↗

Security Vinkius Trust Center ↗

SLA Service Level Agreement ↗

Report Listing Send Report ↗

What your AI can actually do with this

You're tired of clicking through museum websites and digging up dust on digital asset networks just to find one piece of data. This connection lets your AI client talk directly to millions of Smithsonian records, covering everything from scientific artifacts to deep historical finds. You don't have to browse; you just ask.

When you first need to check the archives, you start with the search_records tool. It runs a wide-open query across every single department in the Smithsonian network, letting you plug in general keywords or concepts. This gives you a massive list of potential records when you don't know exactly which area holds what you’re looking for.

If that initial search comes back with too much noise—like finding 'pottery' but needing to filter it down to only the Art department, not Anthropology—you use search_category. This tool lets you narrow your scope by limiting results to specific fields or units. You can run a targeted query for things like 'Art,' 'History,' or 'Science.' It’s how you refine that massive list without having to adjust your initial search terms.

The system handles the full research cycle: first, you run search_records for broad coverage; then, if needed, you use search_category to focus that scope. Once you've filtered down and found a record ID you care about, you need all the details—the provenance, the description, every damn piece of metadata. For that, you call get_content.

This tool pulls out the complete, detailed information payload for one specific object ID within the collection. It gives you everything attached to that item.

When you're building a project, your agent uses these three tools in sequence: run the general search with search_records to get keyword matches; use search_category if those results are too mixed up and need to be limited by discipline; then, take the specific ID that matters most and feed it into get_content to pull all the detailed descriptive data.

You're not just getting a title or a date—you’re pulling the full package for global cultural heritage analysis.

Built · Hosted · Managed by Vinkius Smithsonian Open Access - Museum Records & Data

Server ID 019e38ef-3dda-7245-9cf5-dbe24d318e2d

Vinkius Inspector

Compliance Grade A+

Score 100/100

Report View Report ↗

Here's how it actually works

The bottom line is, you treat the Smithsonian archive like a database connection through your AI client, bypassing manual web scraping entirely.

Subscribe to the server and enter your Smithsonian API Key.

Your AI client executes a tool call (e.g., search_records) based on your prompt's intent, querying the live EDAN network.

The MCP Server returns structured metadata results that your agent processes directly in your environment.

What Changes When You Connect

You don't get generic search results. By using search_records, your agent pulls structured metadata on artifacts, specimens, and historical documents across the entire Smithsonian network.

Instead of guessing where to look, use search_category to filter research immediately by field (e.g., 'American History'). This cuts down irrelevant noise in your results.

When you find a promising ID, don't settle for basic info. Run get_content to pull the full payload—the provenance, detailed description, and asset links—for deep analysis.

The system handles the complexity of multiple data sources. You just ask for 'Apollo 11 records,' and your agent orchestrates the necessary tool calls behind the scenes.

It provides primary source data that's verifiable. This isn't general web scraping; it's programmatic access to established, cataloged collections.

See it in action

01 01

Tracing a specific historical object

A curator needs details on an unknown artifact. They first use search_records with keywords ('Civil War medal'). The results give them IDs. Then, they feed those promising IDs into get_content to pull the full provenance and history of each piece.

02 02

Curating a themed educational module

An educator wants examples of early scientific tools. They use search_category to restrict results to 'Natural History' records, then refine the search within that category until they find enough material for their lesson plan.

03 03

Broad market research on cultural assets

A developer needs a dataset of all art-related items. They start with search_records globally, but immediately narrow the results using search_category to filter only for 'Visual Arts' units before writing any code.

04 04

Verifying source material for an academic paper

A student is researching a specific era. They ask their agent to search records related to the 1920s (search_records), find several IDs, and then use get_content on each one to pull verified metadata payloads for citation.

The honest tradeoffs

Treating it like a search engine

Anti-pattern

Asking the agent simply, 'Tell me about American history.' This yields too much data and doesn't tell you where to find specific assets.

The Fix

Don't ask for general knowledge. First, use search_category to narrow the scope (e.g., 'Art'). Then, use keywords with search_records to focus on a topic within that category.

Missing crucial metadata

Anti-pattern

Getting results from search_records and only seeing titles or dates. This is useless for citation.

The Fix

Always follow up with the IDs you get using get_content. That tool pulls the complete, deep metadata package you actually need.

Ignoring data structure

Anti-pattern

Trying to write a single prompt that does everything. The agent gets confused about whether it should search broadly or filter narrowly.

The Fix

Think in stages: 1) Find the area (search_category). 2) Find the item (search_records). 3) Get the data (get_content). Let your agent execute those steps.

When It Fits, When It Doesn't

Use this server if your task requires verifiable, structured knowledge drawn from major global institutions. The core loop is: Scope -> Search -> Detail. You use search_category when you know the type of data (Art, Science) but not the specific subject. Use search_records when you know the subject (Apollo 11, Impressionism) but need to check multiple departments. Only use get_content once your agent has provided a record ID that needs deep inspection. Don't try to get everything in one go; these tools force you through the proper research pipeline and prevent data fragmentation.

Questions you might have

How do I find a record across all Smithsonian units using search_records? +

You simply instruct your agent to run search_records with your general query. This tool queries the entire EDAN network, giving you results from every department.

Is there a way to filter my search by art or history using search_category? +

Yes, use search_category. You specify the desired unit (like 'Art') and the tool filters all subsequent searches to only show records from that specific taxonomy.

What information does get_content actually retrieve for an ID? +

get_content pulls comprehensive, deep metadata. It includes provenance, detailed descriptive text, and links to the digital assets associated with that record.

Can I combine search_records and search_category in one go? +

While you can't force it into a single tool call, your agent is designed to chain them. You ask for 'Art from the 19th Century,' and the system runs search_category first, followed by a targeted search_records.

How do I properly authenticate when using search_records? +

You must provide your specific Smithsonian API Key in the connection settings. This key authorizes all calls to search_records. Your AI client handles the authentication automatically after you connect it via MCP.

If I use get_content with an invalid record ID, what error message should my AI client expect? +

The server returns a specific 'Record Not Found' code and details. Your agent can check this response to handle missing IDs gracefully without failing the entire process.

Are there limitations on how many records I can query using search_records in a single run? +

The Smithsonian API enforces standard rate limits to prevent overuse. If you hit the limit, your agent will receive an HTTP 429 error and needs to pause before retrying.

Does search_category allow filtering by non-museum data types, like scientific measurements? +

No, search_category maps only to defined Smithsonian collections and units. It restricts search results strictly to established cultural or historical metadata fields for accuracy.

Can I search for specific historical figures across all Smithsonian museums? +

Yes! Use the search_records tool with your query (e.g., 'Abraham Lincoln'). It will return matching records, images, and artifacts from across all Smithsonian units.

How do I get the full metadata for a specific museum object? +

Use the get_content tool with the unique identifier (ID) of the record. This will fetch detailed metadata, including descriptions, dates, and media links.

Is it possible to limit my search to just art or science categories? +

Yes, the search_category tool allows you to specify a category (like 'art', 'history', or 'science') along with your search query to get more targeted results.

Connect to your AI in seconds.

Search category

Get content

Search records

Smithsonian Open Access MCP Server: 3 Tools for Research

Make your AI actually useful.

Search Category

Get Content

Search Records

Security and governance baked right in.

Claude AI

Open Claude Settings

Add Custom Connector

Start a conversation

Claude Code

Open your terminal

Add the MCP Server

Start coding

Cursor

One-Click Install (Recommended)

Open Cursor Settings

Add New Server

Use in Composer

Antigravity

Configure Agent Environment

Bind the Endpoint

Execute

VS Code Copilot

One-Click Install (Recommended)

Open MCP Settings

Add Server Config

Windsurf

One-Click Install (Recommended)

Open Windsurf Settings

Add Server Endpoint

LangChain

Install Dependencies

Connect the Server

CrewAI

Define the Tool

Execute Task

Choose How to Get Started

Build Your Own

Make Your AI Do More

Works with Claude, ChatGPT, Cursor, and more

Manually researching global archives is a nightmare of tabs and redirects.

Smithsonian Open Access MCP Server: Get full artifact metadata.

What your AI can actually do with this

Here's how it actually works

Who is this actually for?

What Changes When You Connect

See it in action

Tracing a specific historical object

Curating a themed educational module

Broad market research on cultural assets

Verifying source material for an academic paper

The honest tradeoffs

Treating it like a search engine

Missing crucial metadata

Ignoring data structure

When It Fits, When It Doesn't

Questions you might have