Concepts

Rate limits

Per key and per company, counted by the minute.

Over a limit, the API answers 429 RATE_LIMITED with a Retry-After header and retryAfter in the body, both in seconds. Wait that long, then send the same request again.

HTTP/1.1 429 Too Many Requests
Retry-After: 12
{ "error": "Too many requests. Try again in 12 seconds.", "code": "RATE_LIMITED", "retryAfter": 12 }

• The company’s limit is shared by all its keys: two systems on one company share 120 a minute.
• Sync by updatedSince rather than reading every record each time (Pagination and sync).
• Fetch PDFs only when someone needs them.