API Reference

Rate limits

Per-token, per-organization, and per-IP sliding windows.

Buckets

Per-token — the most common limit. 100 req/min by default.
Per-organization — protects your workspace from a runaway client. 1,000 req/min by default.
Per-IP — applies to public endpoints (pk keys, application submissions). 30 req/min by default.

Headers

Every response includes RFC 9331-style headers so you can throttle proactively. When you exceed a bucket the API responds with 429 rate_limit_error and a Retry-After value in seconds.

RateLimit-Limit: 100
RateLimit-Remaining: 87
RateLimit-Reset: 42

Custom limits

Higher-volume integrations can request a per-token override from duke@ceevee.cc. We keep limits transparent — overrides show up in the developer dashboard next to the affected key.