ScenebrewDevelopers

Rate limits

No per-key limits are enforced today; here is how they will look when they arrive.

The API does not enforce per-key request limits yet. Fair use applies: AI steps are bounded by your credits, and the platform may shed load with 502 responses under pressure, which you should retry with backoff.

When limits are introduced they will be announced in the changelog ahead of enforcement and will follow the usual shape:

  • a 429 Too Many Requests problem document with a Retry-After header;
  • per-key windows, so one busy integration does not throttle another.

Write clients so that 429 and 502 are retried with exponential backoff and jitter, and nothing else changes when limits switch on.

Pagination

Cursor pagination on every collection — limit, next_cursor, and message paging by id.

API reference

Every operation of api.scenebrew.com, rendered from the published OpenAPI contract.