> ## Documentation Index
> Fetch the complete documentation index at: https://support.configview.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Greenhouse setup

Configview reads your Greenhouse recruiting account through the **Harvest v3 API** (`https://harvest.greenhouse.io/v3`), using an **OAuth client credential** that a Greenhouse Site Admin creates. Configview sends only `GET` requests (plus the token request). It never creates, changes or deletes anything in Greenhouse.

You will end up with **2 required secrets** (`GREENHOUSE_CLIENT_ID`, `GREENHOUSE_CLIENT_SECRET`) in Configview when setup is complete.

> **Scope of this integration today.** Configview reads who has a Greenhouse account and at which level (site admin, job admin roles, deactivated, service account), which user holds which role on which job, the hiring team of each job (hiring manager, recruiter, coordinator, sourcer), jobs with their status and dates, departments and offices, the definitions of your custom fields, and **counts** of applications per job, status and stage. It never stores candidate names, emails, phone numbers, resumes, notes, answers, scorecards, offers or demographic data. It does not read interviews, offers, scorecards, the activity feed or the Greenhouse audit log (see "Known gaps").

***

## Harvest v1/v2 no longer works

Greenhouse removed the Harvest v1/v2 API (Basic auth with an API key) on **August 31, 2026**. An old API key will not work here. Configview uses Harvest v3 with OAuth 2.0 client credentials (`https://auth.greenhouse.io/token`).

***

## What you need

* A Greenhouse account where you are a **Site Admin** (to create API credentials)
* Permission to create credentials under **Configure -> Dev Center -> API Credential Management**
* The service user behind the credential must itself be a **Site Admin**: Greenhouse requires that for every list endpoint

***

## Step 1: Open the Greenhouse page in Configview

Open Configview in a second browser tab and leave it open:

`https://{companyname}.configview.com/admin/integrations/greenhouse`

***

## Step 2: Create the Harvest v3 credential

1. In Greenhouse open **Configure -> Dev Center -> API Credential Management**
2. Click **Create new API credential**
3. Choose **Harvest v3 (OAuth)** and the type **Custom**
4. Name it `Configview`
5. Grant these **read (list) scopes only**:

`harvest:users:list`, `harvest:user_roles:list`, `harvest:user_job_permissions:list`, `harvest:jobs:list`, `harvest:job_hiring_managers:list`, `harvest:job_owners:list`, `harvest:applications:list`, `harvest:departments:list`, `harvest:offices:list`, `harvest:custom_fields:list`

6. Copy the **Client ID** and the **Client Secret**. Greenhouse shows the secret once; rotating it invalidates the old one.
7. In the Configview tab paste them into `GREENHOUSE_CLIENT_ID` and `GREENHOUSE_CLIENT_SECRET` under **Credentials** and click the save icon

***

## Step 3: Connect and verify

1. Click **Connect**. Configview creates its tables and schedules every collector at your default run time. Stop any you don't want under **Collectors**:

| Script | Notes |
| - | - |
| **Users** | Every user, active and deactivated, with site admin flag and service accounts. |
| **Roles and Job Permissions** | Permission roles and who holds which role on which job. |
| **Jobs** | Every job with status, department, offices and dates. Notes are not read. |
| **Job Hiring Teams** | Hiring managers, recruiters, coordinators and sourcers per job, as user ids. |
| **Application Counts** | Applications per job, status and stage. Counts only. Reads every application page, so it is the slowest collector on a large account. |
| **Departments and Offices** | The org structure. |
| **Custom Fields** | Field definitions only, never values. |

2. Click **Verify now**. The health check reads both secrets, requests a token, then makes one tiny request per list.

If a check fails:

* **The token step fails with 400 or 401.** The client ID or secret is wrong, was rotated, or a Harvest v1/v2 API key was pasted.
* **A list fails with 403.** The credential was not granted that scope, or its service user is not a Site Admin.

***

## Data Tables

Each table has a `run_at` column and keeps only the newest run.

| Table | Source | Key Columns |
| - | - | - |
| `greenhouse_users` | `GET /v3/users` | user\_id, name, primary\_email, site\_admin, deactivated, is\_service\_account, office\_ids, department\_ids |
| `greenhouse_user_roles` | `GET /v3/user_roles` | role\_id, name, role\_type |
| `greenhouse_user_job_permissions` | `GET /v3/user_job_permissions` | user\_id, job\_id, role\_id, automated |
| `greenhouse_jobs` | `GET /v3/jobs` | job\_id, name, requisition\_id, status, confidential, opened\_at, closed\_at |
| `greenhouse_job_team` | `GET /v3/job_hiring_managers`, `GET /v3/job_owners` | job\_id, user\_id, team\_role, responsible |
| `greenhouse_application_counts` | `GET /v3/applications` (4 fields) | job\_id, status, stage\_name, application\_count |
| `greenhouse_departments` | `GET /v3/departments` | department\_id, name, parent\_id |
| `greenhouse_offices` | `GET /v3/offices` | office\_id, name, parent\_id, location, primary\_contact\_user\_id |
| `greenhouse_custom_fields` | `GET /v3/custom_fields` | field\_id, name, field\_type, private, active, expose\_in\_job\_board\_api |

***

## Things worth knowing

**No candidate data.** Applications are requested with only `job_id`, `status`, `stage_name`, `prospect` and `created_at`, folded into counts in memory, and never stored per candidate. Users and jobs are requested with an explicit field list that leaves out custom field values and notes. Linked candidate records on a user are only counted.

**Hiring team seats outlive people.** A deactivated user can still be named on an open job's hiring team. Configview flags it.

**Rate limits.** Harvest v3 answers `429` with a `Retry-After` header; the numeric limit was not published on the pages Configview was built from (v1/v2 allowed 50 requests per 10 seconds). Configview paces itself at two requests a second and reads lists 500 rows at a time.

***

## Known gaps

* **Audit log.** Greenhouse's Audit Log API (`auditlog.us.greenhouse.io`) is an extra feature that Greenhouse must enable for your account, covers only the last 30 days, and documents a token exchange that needs a Harvest v1/v2 API key. Until Greenhouse documents v3 access for it, Configview does not read it.
* **Approvals, offers, scorecards, interviews and candidates** are deliberately not collected.
* **First-tenant checks.** Field names and list paths come from Greenhouse's v3 reference, not a live account. Expect small corrections after the first real run (for example, whether `/v3/users` returns deactivated users by default, and the exact list paths for hiring managers and job owners).


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.