Greenhouse Harvest API

What authentication does the Greenhouse Harvest API use?

The API uses HTTP Basic authentication with your Harvest API token as the username and an empty password. Tokens are generated per-user in Greenhouse settings with configurable permission scopes. Through Jentic, the Harvest API token is stored in your Jentic One instance so agents authenticate without handling the raw token.

Can I advance or reject applications through the Greenhouse Harvest API?

Yes. POST /applications/{id}/advance moves an application to the next configured stage in the job's interview plan. POST /applications/{id}/reject marks an application as rejected with an optional rejection reason ID and email notification flag. Both operations require the application ID and appropriate API token permissions.

What are the rate limits for the Greenhouse Harvest API?

Greenhouse enforces a rate limit of 50 requests per 10 seconds for the Harvest API. Exceeding this returns HTTP 429 with a Retry-After header. List endpoints support pagination via the per_page parameter (max 500) and Link headers for cursor-based traversal of large result sets.

How do I retrieve interview scorecards through Jentic?

Search Jentic for 'get interview scorecards for candidate', load the schema for GET /scorecards with the application_id filter parameter, and execute. The response includes each scorecard's overall recommendation, attribute ratings, and interviewer comments. Install with pip install jentic and authenticate through Jentic One, the self-hosted execution layer.

What candidate data can I access through the Harvest API?

GET /candidates returns candidate profiles including name, email addresses, phone numbers, tags, and linked applications. GET /candidates/{id} provides full detail for a single candidate. The applications sub-resource shows which jobs they applied to, their current stage, and rejection status if applicable.

Can I limit what my agent is allowed to do with the Greenhouse Harvest API?

Yes. Jentic One is self-hosted, so you run it and your own rules decide which Greenhouse Harvest operations and credentials your agent may use. Because the application id sits in the URL path, such as /applications/{id}/advance and /applications/{id}/reject, a rule can pin the agent to a single application and let it read that application's stage and nothing else. Write operations like advancing or rejecting an application are only callable if you explicitly add them, so a read-only agent stays limited to endpoints like GET /candidates and GET /scorecards.