Conventions
Rate limits
120 calls and 30 writes a minute, per token.
Limits are per token, so one busy integration cannot starve another. Both REST and MCP calls count.
| Budget | Limit | Window |
|---|---|---|
| All calls | 120 | per minute |
| Writes | 30 | per minute, on top of the call budget |
When you hit one
You get 429 and a Retry-After header in seconds. Wait that long, then continue. A refused write still spends its quota, so retrying in a tight loop only extends the wait.
response
HTTP/1.1 429 Too Many Requests
Retry-After: 17
{ "error": "Slow down: this token has used its 30 writes for this minute." }