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

# Sandbox

> Test supported API endpoints without mutating production data

## How it works

Use the sandbox to test API calls before switching to production. Your organization API key is
available in [Organisation settings](https://app.collective.work/recruiter/organization/api).
When you're ready, request production API access at
[support@collective.work](mailto:support@collective.work).

> [!NOTE]
>
> The [Analytics sandbox](https://docs.collective.work/docs/analytics.md#sandbox) returns fictional data with a valid organization API
> key. Production API access and analytics access are not required; the organization must have at least
> one recruiter.

* The jobs you create in the sandbox are no visible on the app, they are purely internal and accessible only by API,
  for testing purpose.
* For talent import, it is the same and we just do validation of your data. We don't store it.

## Endpoint

Sandbox endpoint is available at:

[https://api.collective.work/sandbox/v1](https://api.collective.work/sandbox/v1)

Whereas the production endpoint is available at:

[https://api.collective.work/production/v1](https://api.collective.work/production/v1)

## Testing

To test the API, use your API key and hit **[Try it](https://docs.collective.work/api-reference/endpoint/job/create.md)** on the documentation for API endpoint

![try-api](https://mintcdn.com/collective/NltUmf_eeFge57l7/images/try-it.png?fit=max\&auto=format\&n=NltUmf_eeFge57l7\&q=85\&s=4947a227371e1cc19c2c3b17ab9b13ba)

![try-api-post](https://mintcdn.com/collective/NltUmf_eeFge57l7/images/try-it-post.png?fit=max\&auto=format\&n=NltUmf_eeFge57l7\&q=85\&s=de6e2dbeb40096a3352db35ef7b4539b)

## Analytics sandbox

Call `GET https://api.collective.work/sandbox/v1/analytics` with your organization API key in the
`Authorization: Bearer <organization API key>` header. The response has the same structure as production,
using four fixed fictional recruiters shared across accounts, including one with zero activity.

Start without `recruiterIds` to discover their IDs, names, and `example.com` emails in `recruiters`.
Filter subsequent sandbox requests using those fictional IDs; real organization recruiter IDs are
rejected. Date ranges, breakdowns, selected sections, and validation follow the same rules as production.

The sandbox generates data in memory and does not query your organization's analytics. Counts for the
same fictional recruiters and resolved UTC dates stay consistent across requests; `metadata.requestedAt`
changes, and `metadata.organizationId` identifies your authenticated organization. Preset date windows
move with the current UTC date. See the [Analytics guide](https://docs.collective.work/docs/analytics.md#sandbox) for curl and JavaScript
examples, or use [Get Analytics](https://docs.collective.work/api-reference/endpoint/analytics/get.md) with the sandbox environment.

## Note on sandbox data persistence

Stored sandbox data, such as test jobs, is ephemeral and will not be persistent forever.
We reserve ourselves the right to clean the data from time to time during the weekend.
Analytics sandbox data is generated for each request and does not depend on stored test jobs or activity.


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