canonical: https://jentic.com/apis/artic.edu/artic # Artic Edu Art Institute of Chicago API The Art Institute of Chicago API is a public, read-only aggregator over the museum's collections data. It exposes 97 endpoints across artworks, artists, agents, places, galleries, exhibitions, sites, events, articles, tours, mobile sounds, publications, sections, products, images, videos, and sounds. Each primary entity supports list and detail operations, and 28 of the 34 resource groups also expose a /search operation, including artworks, agents, places, galleries, exhibitions, events, and articles. Artists are listed and fetched by id only; artist search runs through /agents/search, since the spec files the artist operations under its agents tag. The API is unauthenticated, intended for educational and research use, and serves as an open data layer for museum integrations. This entry is built from a third-party copy of the museum's OpenAPI document, pinned at version 1.10 and generated in August 2024; the museum publishes its own OpenAPI file at api.artic.edu/api/v1/openapi.json, which is the canonical description and lists 103 operations as of its version 1.14. ## For AI agents Search and fetch artworks, artists, exhibitions, and related collections data from the Art Institute of Chicago - no authentication required. ## Scope Does not handle ticketing, membership, or e-commerce - use for reading the Art Institute of Chicago's open collection and editorial data only. ## Capabilities - Search the artwork collection by query string and filters - Fetch artwork detail records including titles, artists, dates, and images - Search and fetch agents, places, galleries, and exhibitions, and list or fetch artist records - Browse curated tours, mobile audio guides, and educational articles - Retrieve publications, sections, and product records - Look up image, video, and sound assets associated with collection records ## Use cases ### Educational App Integration Power a museum education app or classroom tool by searching the Art Institute of Chicago collection through /artworks/search and rendering details from /artworks/{id}. Because the API is unauthenticated, classroom deployments avoid credential management entirely. Strong fit for art history coursework that needs licensed-clean reproductions and curator-written descriptions. Example prompt: Call /artworks/search with q='impressionism', then for each hit call /artworks/{id} and render title, artist, and primary image url ### Cultural Discovery Feed Build a daily discovery feed for design or art newsletters by pulling new exhibitions, articles, and featured artworks. /exhibitions and /articles surface what's currently on at the museum, while /artworks/search with an Elasticsearch domain syntax query on date_start can rotate featured pieces. The unauthenticated read model makes it well suited to static site builds. Example prompt: GET /exhibitions, keep the records whose status field reads open, take the top 5, and write them as JSON for a newsletter generator ### Tour and Audio Guide Surfacing Surface museum tours and mobile audio guide content inside a third-party visitor app. /tours and /mobile-sounds expose curated walkthroughs and narration, so partner apps don't need to redesign tour content from scratch. Example prompt: Fetch /tours, take the first tour, and for each stop GET /artworks/{id} plus /mobile-sounds for the narration audio ### Research Catalog Mining Mine the museum's public catalog for academic research - for example, computing how the artwork collection's coverage of a movement or geography has shifted over time. The /artworks, /artists, /places, and /agents endpoints provide enough metadata to build datasets without scraping HTML. Example prompt: Query /artworks/search with an Elasticsearch domain syntax query on place_of_origin, aggregate counts by century from date_start, and emit a CSV of works-per-century for that place ### Agent-Driven Art Lookup via Jentic Give an AI assistant a tool that answers art-history questions by calling the Art Institute of Chicago API. The agent searches Jentic for 'search artworks at Art Institute of Chicago', loads the schema for /artworks/search, and executes - with no credentials to manage. Example prompt: Search Jentic for 'search artworks at Art Institute of Chicago', load the schema, and execute with q='Monet water lilies' to retrieve matching artworks ## Key endpoints | Method | Path | Description | | --- | --- | --- | | GET | /artworks | List artworks | | GET | /artworks/search | Search artworks by query and filters | | GET | /artworks/{id} | Fetch a single artwork | | GET | /artists | List artists | | GET | /exhibitions | List exhibitions | | GET | /galleries/search | Search galleries | | GET | /tours | List curated tours | | GET | /articles | List published articles | ## Key resources - **Artworks** — List, search, and fetch artwork records and images - **Artists** — List and fetch artist records and their links to works; artist search runs through /agents/search - **Agents** — Broader people/organisation records that include artists, donors, and collectors - **Places** — List, search, and fetch geographic places associated with works - **Galleries** — List and fetch museum galleries - **Exhibitions** — List and fetch museum exhibitions - **Tours and Mobile Sounds** — Curated tour content and audio guide narrations - **Articles, Publications, Sections** — Editorial content published by the museum - **Images, Videos, Sounds** — Media assets associated with collection records ## Why Jentic - **Setup:** Wiring the Art Institute of Chicago API by hand means resolving its host and picking the right path from roughly a hundred collection and editorial endpoints, even with no credential to manage. Through Jentic you install once, import the Art Institute of Chicago API from the API Directory, and your agent calls it with no key to configure. - **Permission scoping:** This API exposes only GET reads over open collection and editorial data, so a rule can limit the agent to the operations it needs, such as searching artworks or listing exhibitions. You choose the operations it may call, and no write operation exists to include. - **Credential handling:** The Art Institute of Chicago API is unauthenticated, so there is no credential to store. Jentic still routes the call through your own Jentic One instance at execution time for consistent handling alongside authenticated APIs. - **Discovery method:** Agents search Jentic by intent such as 'search artworks at the Art Institute of Chicago' or 'list exhibitions', and Jentic returns the matching operation with its input schema so the agent calls the right endpoint without reading the reference docs. ## Related APIs - **Harvard Art Museums API** — Harvard Art Museums offers a similar open-access museum collection API - **Cooper Hewitt API** — Cooper Hewitt is the Smithsonian's design museum, exposing a comparable open collection API - **Figshare API** — Figshare hosts academic datasets that can be linked from museum-collection research outputs ## FAQ ### What authentication does the Art Institute of Chicago API require? None for read access. The API is public and unauthenticated for collection browsing and search: all 97 operations declare no security scheme. The museum's docs ask you to add an AIC-User-Agent header naming your project and a contact email, but no API key is required. Through Jentic there is no credential to store for this API. ### Can I search for artworks by an artist or movement? Yes. GET /artworks/search accepts a q parameter for free-text search. For field-level filters such as artist_id or a date_start range, pass an Elasticsearch domain syntax query through the query parameter, which the spec provides for exactly that purpose. Resolve artist ids first with /agents/search, since the API files artists under its agents tag and /artists itself has no search operation. ### What are the rate limits for the Art Institute of Chicago API? The museum's API documentation states that anonymous users are throttled to 60 requests per minute, with each IP counted as a separate user. It also asks you to send an AIC-User-Agent header naming your project and a contact email. Cache responses and avoid per-request calls in user-facing loops to stay inside that throttle. ### How do I get the image for an artwork? Each artwork record includes an image_id which can be combined with the IIIF image base URL to produce an image. /images and /images/{id} also expose image asset records directly when you need additional metadata. ### How do I search the collection through Jentic? Install Jentic One on the machine that will host it, run jentic register on the separate machine where your agent runs, then import the Art Institute of Chicago API from the API Directory. Your agent searches Jentic for 'search artworks at the Art Institute of Chicago', loads the schema for GET /artworks/search, and executes with your query string. That operation pages with its from and size parameters, which the spec describes as Elasticsearch conventions. ### Can I list current exhibitions at the museum? Yes. GET /exhibitions returns every exhibition with a status field recording whether it is open or closed, plus aic_start_at and aic_end_at dates. The list operation declares no query filters, so select the open records yourself, or use GET /exhibitions/search with the query parameter. ### Is there an Art Institute of Chicago MCP server? You don't need an MCP server to give your agent the Art Institute of Chicago API. Jentic connects it directly from the API Directory: import it and your agent gets the full 97-endpoint surface, so it can search artworks, browse exhibitions, and pull tour audio without another server's tool definitions loaded into its context. Because the API is unauthenticated you have no credential to configure either, and your own rules still decide which of the 97 operations the agent may call. ### Can I limit what my agent is allowed to do with the Art Institute of Chicago API? Yes. Because you run Jentic One yourself, your own rules decide which operations the agent may call, so you can restrict it to just the endpoints it needs, such as searching artworks via GET /artworks/search or listing exhibitions via GET /exhibitions. This API exposes only GET reads over open collection and editorial data, so there are no write operations to include and the agent cannot alter anything at the museum. Since the API is unauthenticated, there is no credential to hand out, and the call still routes through your own self-hosted instance so scoping is handled the same way as for authenticated APIs.