Install Jentic One Beta
Jentic One is a self-hosted execution layer for AI agents. It lets your agent call the NPR Listening Service, or any other public or private API you need. You set the rules, the agent never sees your credentials, and every call is logged.
Two steps, two machines. Install the instance in a safe environment, then register your agent from wherever it runs.
Step 1: Jentic One Host machine
# On the machine that will host your Jentic One instance:
curl -fsSL "https://jentic.com/install.sh?src=apis&api=%2Fapis%2Fnpr.org%2Fnpr-listening" | shStep 2: Agent machine
# On the machine where your agent runs (keep this separate from the instance):
curl -fsSL "https://jentic.com/install.sh?src=apis&api=%2Fapis%2Fnpr.org%2Fnpr-listening" | sh
jentic register # connects your agent to your Jentic One instanceJentic One is in public beta. The setup above keeps your agent separate from the instance, which is what you want before using real credentials: an agent running as the same OS user as Jentic One can read its stored keys directly. Just evaluating? A single local install is fine to start. See the secure deployment guide for the tiers.
What an agent can do with NPR Listening Service API.
Retrieve a personalised list of audio recommendations for the logged-in listener
Submit listener ratings to record completion, skip, or share events that shape future recommendations
Search for audio and aggregations matching a query, returning ranked recommendation items
List the available NPR One listening channels for a listener
GET STARTED
Fetch recommendations scoped to a specific NPR member organisation or category
Retrieve recent listener history including the latest promo audio heard
Patterns agents use NPR Listening Service API for, with concrete tasks.
★ Personalised Audio Stream for NPR One Apps
An NPR One client app calls GET /v2/recommendations on launch to receive a queue of audio segments tailored to the listener - newscasts first, then stories, podcasts, and station promos blended by the recommendation engine. As each item plays, the app submits a rating with POST /v2/ratings so the next call returns a refined queue.
Call GET /v2/recommendations with the listener's access token, then submit POST /v2/ratings with completion=true after each item plays to the end.
Search Across NPR Audio
Listeners often want to jump to a specific topic - election coverage, a favorite show, an interview they remember. GET /v2/search/recommendations returns a ranked list of matching audio and aggregation items, which client apps render as a search results page with playback hooks back into the recommendation queue.
Call GET /v2/search/recommendations with q=climate policy and the listener's access token, then play the top-ranked item on the listener's device.
Local Station Discovery via Organisations
Listeners often want a feed scoped to their local NPR member station, optionally filtered by category. GET /v2/organizations/{orgId}/recommendations returns a feed for that station, and GET /v2/organizations/{orgId}/categories/{category}/recommendations narrows it to news, arts, or another category. This powers the local view inside NPR One client apps.
Call GET /v2/organizations/{orgId}/categories/news/recommendations with the listener's home station orgId to render a local-news feed.
Agent-Driven Voice Listening Session
A voice agent on a smart speaker fetches the next NPR One recommendation, plays it, and reports the completion event back through Jentic. Because Jentic exposes both the recommendation fetch and the rating post as discrete operations, the agent can handle the full listening loop without bespoke integration code.
Search Jentic for 'get the next npr audio recommendation', execute GET /v2/recommendations, play the first item, then call POST /v2/ratings with the rating event.
9 endpoints — jentic publishes the only available openapi specification for npr listening service, keeping it validated and agent-ready.
METHOD
PATH
DESCRIPTION
/v2/recommendations
Get the listener's personalised recommendation queue
/v2/ratings
Submit a rating to shape future recommendations
/v2/search/recommendations
Search NPR audio and aggregations
/v2/channels
List the listener's available channels
/v2/organizations/{orgId}/recommendations
Get recommendations scoped to a specific organisation
/v2/organizations/{orgId}/categories/{category}/recommendations
Get recommendations for a category from a specific organisation
/v2/aggregation/{aggId}/recommendations
Get recommendations from a specific aggregation independent of the personalised queue
/v2/history
Get recent ratings the listener has submitted
/v2/recommendations
Get the listener's personalised recommendation queue
/v2/ratings
Submit a rating to shape future recommendations
/v2/search/recommendations
Search NPR audio and aggregations
/v2/channels
List the listener's available channels
/v2/organizations/{orgId}/recommendations
Get recommendations scoped to a specific organisation
/v2/organizations/{orgId}/categories/{category}/recommendations
Get recommendations for a category from a specific organisation
/v2/aggregation/{aggId}/recommendations
Get recommendations from a specific aggregation independent of the personalised queue
What agents get from Jentic-routed access to this vendor.
Setup
Wiring the NPR Listening Service by hand means carrying an OAuth 2.0 bearer token into every call to listening.api.npr.org and hand-mapping recommendations, search, ratings, and history yourself. Through Jentic you install once, import the NPR Listening Service from the API Directory, store the access token once, and your agent calls it.
Permission scoping
This service puts the organization and aggregation ids in the URL path (/v2/organizations/{orgId}/recommendations), so a rule can pin your agent to one organization's recommendations and nothing else. You choose the operations it may call, so writing ratings is not included unless you add it.
Credential isolation
Your NPR OAuth 2.0 access token is stored once, encrypted, by your own Jentic One instance and injected into the Authorization header at execution time. It never enters the agent's prompt, logs, or context.
Intent-based discovery
Agents search Jentic by intent such as 'get the next NPR audio recommendation' or 'search recommendations by keyword', and Jentic returns the matching Listening operation with its input schema so the agent calls the right endpoint without browsing the reference docs.
Alternatives and complements available in the Jentic catalogue.
Specific to using NPR Listening Service API through Jentic.
Why is there no official OpenAPI spec for NPR Listening Service?
NPR does not publish an OpenAPI specification for the Listening Service. Jentic generates and maintains this spec so that AI agents and developers can call NPR Listening Service via structured tooling. It is validated against the live API and kept up to date. Get started with Jentic One, the self-hosted execution layer.
What authentication does the NPR Listening Service use?
Every endpoint requires an OAuth 2.0 bearer access token issued by the NPR Authorization Service. Jentic holds the listener's token in its encrypted vault and adds the Authorization header on each call, so agents never see the raw token.
Can I get personalised audio recommendations with the NPR Listening Service?
Yes. GET /v2/recommendations returns a personalised queue for the logged-in listener, and submitting feedback to POST /v2/ratings refines future calls. The same engine also powers organisation-scoped and category-scoped feeds via /v2/organizations/{orgId}/recommendations.
How do I record a listener rating through Jentic?
Search Jentic with 'submit npr rating', load POST /v2/ratings, and execute it with the rating event (completion, skip, share, or thumbs) along with the media id and the listener's access token. The next /v2/recommendations call reflects the new feedback.
Does the NPR Listening Service support search?
Yes. GET /v2/search/recommendations accepts a query string and returns a ranked list of audio and aggregation items, which is how NPR One client apps build their topic-search experience.
Can I limit what my agent is allowed to do with the NPR Listening Service?
Yes. Because you run Jentic One yourself, your own rules decide which NPR Listening operations and which stored OAuth token your agent may use. Since the organization and aggregation ids sit in the URL path, such as /v2/organizations/{orgId}/recommendations, you can pin the agent to a single station's recommendations and nothing else. You also choose the operations it may call, so writing ratings through POST /v2/ratings is not included unless you add it, letting you keep the agent read-only over recommendations, search, and history.
Know of an official OpenAPI document? Contribute it →
For Agents
Fetch personalized NPR audio recommendations, search for episodes, list channels, and submit listener ratings to shape future recommendations.
Use for: I need to get the next audio recommendation for the logged-in NPR One listener, Submit a rating after the listener finished an episode, Search NPR One for audio about climate policy, List the available NPR One channels
Not supported: Does not handle OAuth token issuance, listener profile management, or sponsorship - use for personalised NPR One audio recommendations, search, ratings, and history only.
Jentic publishes the only available OpenAPI specification for NPR Listening Service, keeping it validated and agent-ready. The NPR Listening Service delivers personalised audio recommendations to NPR One client apps. It exposes the main recommendation feed for a logged-in listener, channel listings, search-driven recommendations, organisation-scoped feeds, aggregation feeds, listening history, and a ratings endpoint that records listener feedback to shape future recommendations. All endpoints require an OAuth 2.0 access token issued by the NPR Authorization Service.
/v2/history
Get recent ratings the listener has submitted