Rate limits & errors

Rate limits

Endpoints are rate-limited per account. Organization endpoints allow 60 requests per minute; some sensitive operations carry tighter, per-endpoint limits. Exceeding a limit returns HTTP 429 with a structured body:

{
  "status": "error",
  "message": "Too many requests",
  "errors": [
    {
      "type": "RateLimitError",
      "key": "rate_limit",
      "retry_after": 42
    }
  ]
}

Honor retry_after (seconds) before retrying. For bulk work, spread writes instead of bursting — limits are per minute, so a steady 1 req/s never throttles.

Status codes

Code	Meaning	Typical cause
200 / 201	OK / created	—
400	Bad request	Malformed body, token scope not valid for the route
401	Unauthenticated	Missing, revoked, or expired API key
403	Forbidden	Route not available to API keys (credential/billing), plan gate, org role
404	Not found	Wrong id — or a resource your role can’t see (Lynqu never confirms existence of records you can’t access)
422	Validation failed	Field errors listed in errors[]
429	Rate limited	Retry after retry_after seconds
5xx	Server error	Safe to retry with backoff

Error envelope

errors[] entries are machine-readable:

{
  "type": "ValidationError",
  "key": "label",
  "errorMessage": "The label field is required.",
  "errorTitle": "Validation failed"
}

Branch on type and key; errorMessage/errorTitle are localized, human-readable strings — display them, don’t parse them.