Errors & rate limits

Error response

All error responses use this JSON shape:

{
  "error": {
    "type": "authentication_error",
    "code": "invalid_auth",
    "message": "Human-readable explanation.",
    "request_id": "unique-id-for-support"
  }
}

Common type values:

`type`	HTTP	Meaning
`invalid_request`	400	Malformed input
`authentication_error`	401	Missing/invalid Bearer token
`permission_error`	403	Forbidden or insufficient scope
`not_found`	404	Resource missing
`conflict_error`	409	Conflict (e.g. idempotency)
`validation_error`	422	Validation failed
`rate_limit_error`	429	Too many requests
`billing_error`	402	Billing / entitlement
`internal_error`	500	Unexpected server error

Include request_id when contacting support.

Rate limiting

When Upstash Redis is configured:

Auth	Window	Limit
Session JWT	1 minute	~100 requests
API key	1 minute	~60 requests

If Redis env vars are not set, limits may be disabled in development.

Response 429

{
  "error": {
    "type": "rate_limit_error",
    "code": "rate_limit_exceeded",
    "message": "Too many requests. Please try again later.",
    "request_id": "req_01abc..."
  }
}