Rate limits

Rate limits are enforced at the Kong gateway in front of every Commotion API. Limits apply per API key and reset every minute.

Per-plan limits

Plan	Requests per minute	Concurrent streams
Developer	60	2
Growth	300	10
Scale	1,500	50
Enterprise	Custom	Custom

Sandbox keys (eak_test_*) inherit the rate limit of the plan they belong to.

429 response shape

When a request is rate-limited, the gateway returns:

HTTP/1.1 429 Too Many Requests
Retry-After: 17
Content-Type: application/json

{
  "error": "rate_limited",
  "message": "Too many requests. Retry after 17 seconds.",
  "retry_after": 17
}

Retry-After is in seconds and is always present on 429 responses.

Recommended client behaviour

• Respect Retry-After. Sleep for the indicated duration before retrying.
• Add a small jitter (50 to 250 ms) when retrying multiple in-flight requests, to avoid herding.
• For long-running batch workloads, throttle client-side rather than relying on 429s as a signalling mechanism.

Tip

Streaming endpoints have a separate concurrent-stream limit. Hitting it returns 429 on the WebSocket upgrade rather than after the connection opens.

Requesting a higher limit

Contact support@gocommotion.com with your account ID, the API in question, and the target limit. Custom limits are available on Scale and Enterprise plans.