Rate limits
| What | Limit |
|---|---|
| Each API key | 120 requests per minute |
| Each dashboard session | 600 requests per minute |
| Attempts with invalid keys, per IP | 30 per minute |
Session routes (/api/auth/*) |
Per-IP limit |
| Sign-ups with the same email | 5 per hour (with confirmation) |
| Resending the confirmation link | 3 per email per hour; then 429 demasiados_reenvios |
| Confirmation links when signing in unconfirmed | 6 per email per hour (counting sign-up ones) |
| Password-reset links | 3 per email per hour; after that none are sent (the response doesn’t change) |
| Password attempts when confirming the email | 5 per account per hour; 10 every 5 minutes per IP |
| Organizations a person owns | 10 |
| Body size | 512 KB |
On a self-hosted installation, the per-key limit is changed with the LIMITE_CLAVE_POR_MINUTO
variable.
Headers
Section titled “Headers”Every response authenticated with an API key includes:
RateLimit-Limit: 120RateLimit-Remaining: 117RateLimit-Reset: 42RateLimit-Reset is the number of seconds until the window resets.
When you exceed the limit
Section titled “When you exceed the limit”429 limite_excedido with the Retry-After header (seconds) and details.reintentarEn:
{ "error": { "code": "limite_excedido", "message": "…", "details": { "reintentarEn": 18 } } }Wait Retry-After seconds and retry. If it was an issuance, use the same
Idempotency-Key. The CLI exits with code 4 and prints the
Retry-After.
A body larger than 512 KB gets 413 with the code datos_invalidos.
For AI assistants: llms.txt ·llms-full.txt ·every page as Markdown by adding .md to its path.