API
Rate limits
Budgets are per key and per minute, with separate windows for reads and writes.
Each API key has a per-minute budget set by the workspace's plan. GET /me reports it as limits.apiRequestsPerMinute. Writes (POST, PATCH, DELETE) draw from a separate budget a third the size, so a burst of reads never blocks a needed update, and the other way round.
Every response carries the state of the budget the request used:
| Header | Meaning |
|---|---|
X-RateLimit-Limit | Requests allowed per minute for this key and kind |
X-RateLimit-Remaining | Requests left in the current window |
X-RateLimit-Reset | Unix time (seconds) when the window resets |
When the budget is spent the API answers 429 rate_limited with a Retry-After header in seconds. Wait that long and retry; do not tighten the loop.
Bulk creation has its own limit of ten batches per ten minutes per key, on top of the write budget.