Smithsonian Open Access MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| SMITHSONIAN_API_KEY | Yes | Your API key from api.data.gov (required for accessing Smithsonian collections) |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
| logging | {} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| search_objectsA | Search Smithsonian collection objects by keyword and filters. Returns compact summaries; pass an id to get_object for the full record. For "how many" questions use limit=1 and read total_count. For works by a person use maker, since query also matches records that only mention them. |
| get_objectA | Full record of one object: description, notes, materials, topics, rights, up to 10 images and the web_url of its page. |
| list_museumsA | Smithsonian units in Open Access: codes, names, the record_type values that return their records, and the names the museum argument accepts. For counts use get_collection_stats. |
| explore_topicA | A varied sample of objects about a topic, spread across museums in proportion to their matches and across object types, with counts by museum and type. Prefers objects with images whose title, type or subject names the topic. Use it for open-ended browsing; use search_objects to find specific things. Calls can return different samples. |
| get_collection_statsA | Counts of searchable objects, archive records, objects with images and objects with CC0 media, for the whole collection or one museum. Each count equals the total_count of the matching search_objects call. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| collection_research | Research a topic across the Smithsonian collections. |
| object_analysis | Analyze one collection object in depth. |
| exhibition_planning | Plan an exhibition from Smithsonian objects. |
| educational_content | Create a lesson built around Smithsonian objects. |
| museum_on_view | Find out what is on display at a museum now. |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| museums | Smithsonian units with codes, record types and accepted name aliases. |
TDQS
Scored across 5 tools
Each tool has a distinct primary purpose: search_objects for specific queries, get_object for full records, list_museums for museum metadata, explore_topic for open-ended browsing, and get_collection_stats for aggregate counts. However, search_objects also advertises using it for 'how many' questions via total_count, which overlaps with get_collection_stats and could cause misselection for count queries.
All tool names follow a consistent snake_case verb_noun pattern (search_objects, get_object, list_museums, explore_topic, get_collection_stats). The verbs are descriptive and predictable, making the set easy to scan.
With only 5 tools, the set is well-scoped for the Smithsonian Open Access domain. Each tool covers a distinct access pattern (search, retrieve, list, explore, aggregate) without redundancy, making it easy to navigate.
The tools cover core workflows: searching, retrieving full records, listing museums, browsing topics, and getting statistics. Minor gaps exist—such as no dedicated tool to list available topics or object types, and no direct media download—but these can be worked around via existing tools.