Reliability

Rate Limits

Design integrations to operate safely when request limits are reached.

Current documented behavior

Protected API keys may return HTTP 429 Too Many Requests after reaching the configured daily allowance.

{
  "success": false,
  "error": {
    "code": "rate_limited",
    "message": "Daily request limit reached. Try again tomorrow."
  }
}

Client recommendations

• Cache public catalog data where appropriate.
• Avoid unnecessary polling.
• Use exponential backoff with jitter for retryable failures.
• Do not automatically retry validation or authentication failures.
• Respect any future Retry-After guidance returned by the API.
• Spread non-urgent jobs over time instead of sending bursts.

Example backoff sequence

1s → 2s → 4s → 8s → 16s