How to Use the API

Access methods

The API is public and available for GET requests at

Get started with your favorite API tool or script. We do not require an API key.

Quick start

The best way to learn what is available via the API is to start making API requests and exploring what comes back, whether in the browsable API with URLs or using another interface. Data is accessed through GET requests using API endpoints. For a list of available endpoints, see the endpoints and parameters section of this document.

This section includes some basic examples to show you how to construct queries and start exploring.

Example 1: Get all collections

Use the /collections endpoint to get a list of all of our archival collections, which are intellectually significant groups of records:


Example 2: Get a specific collection

Get data about one specific collection, replacing {id} with the collection’s identifier (example id: H45i6yf7MUHuaRwQVupvg5). Collection identifiers can be found in the /collections or /search endpoints as the URI value of a collection. They can also be found in DIMES URLs, since DIMES uses the API. For example:


See what a response looks like in the browsable API by opening the link in your browser:

Example 3: Get objects that are available to view online

Use the /objects endpoint and online parameter to get data for all archival objects that have digital versions available. Objects are defined as intellectually significant groups of records in a collection that do not have children:


Note that online is a query parameter. By convention, these are included after a ? in the URL. To add multiple query parameters, separate the parameters by &. For example, to use both the online and start_date query parameters:

Use the /search endpoint to return the number of search matches for the query term “agriculture” that are in collections and have been categorized as photographs with dates between 1940 and 1950:

Note: As documented in the parameters section, appending __gte and __lte to the date parameters function as greater than or equal to and less than or equal to, allowing us to include any start and end dates in this decade instead of limiting ourselves to specific start and end dates.

Example 5: Minimap

Use the /minimap endpoint to return collections and objects with search hits for the query term “agriculture” within the Ford Foundation records collection:

Understanding the data that comes back

We provide JSON-formatted data.

Endpoints are paginated. We show 50 records per page by default. Pagination can be controlled via the following query parameters:

  • limit to set how many records each page will return
  • offset to request a specific page of results.

Example: get the second page of agent records consisting of 10 records:


Using the API client with Python

Example Python scripts that uses the RAC API client:

Example 1: Size of collection

Find out the physical size (called extent) of the Social Science Research Council records collection in the archives.

# import rac_api_client module
from rac_api_client import Client

# get the collection data about the Social Science Research Council 
# records using the collection id.
client = Client()
collection = client.get("/collections/iNo7dbyWw2GwSwKsC3nDj3")

# print collection title

# print collection extent value and type to get size
for extent in collection["extents"]:
    print(extent["value"], extent["type"])


Social Science Research Council records
509.06 Cubic Feet

Example 2: Collection creators

Identify the creators of collections that contain keyword search matches for “public television”. The /search endpoint performs search queries across agents, collections, objects, and terms.

Creators are the people, organizations, or families responsible for creating the records. Terms are controlled values describing topics, geographic places, or record formats.

# import rac_api_client module
from rac_api_client import Client

# create an empty set of creators (people or organization) of collections 
# that contain search matches for the query
creator_set = set()

# search across agents, collections, objects and terms for "public television"
# add the associated creators to the creator set
client = Client()
for records in client.get_paged("/search", params={"query": "public television"}):

# convert the set to a list of creators with no duplicated names
dedup_creator_list = list(creator_set)

# print deduplicated list of creators


['Rockefeller, David (1915-2017)', 'Knowles, John H. (1926-1979)', 'Ford Foundation', 'Reich, Cary', 'Rockefeller Foundation', 'John and Mary R. Markle Foundation', 'Rockefeller, Nelson A. (Nelson Aldrich)', 'Foundation for Child Development', 'Linden, Patricia', 'Asian Cultural Council', 'Rockefeller, John D., III (John Davison), 1906-1978', 'Rockefeller, Laurance Spelman', 'Arts in Education Program (U.S.)', 'National Committee on United States-China Relations', 'JDR 3rd Fund', 'Downtown Lower Manhattan Association', 'Henry Luce Foundation', 'Rockefeller, John D., Jr. (John Davison), 1874-1960', 'Grant, W. T. (William Thomas)', 'Knight Foundation', 'William T. Grant Foundation']