Guides
Rate limits
Creating confirmations is limited to 30 an hour per client address in the current release. Reads, advance and cancel aren't limited.
Creating a confirmation sends real emails, texts and calls to real people, so it's the one thing the API limits. Everything else is unlimited in the current release.
Current limits
| Endpoint | Limit |
|---|---|
POST /v1/confirmations | 30 requests in any rolling hour, per client IP address |
GET /v1/confirmations, GET /v1/confirmations/{id} | Not limited |
POST …/advance, POST …/cancel | Not limited |
Requests over the limit get a 429 and nothing is created.
The window slides: each create counts for one hour from when it was made, so capacity comes back gradually rather than all at once.
Note
These limits suit a pilot. When you move to production volumes, we raise them for your account. Tell us your peak hour (for example, the evening batch from your order system) and we'll size it.
Handling 429
- Spread batch creates over time instead of sending a whole day's jobs in one burst.
- On a
429, back off and retry later. There's noRetry-Afterheader yet, so wait at least a few minutes before trying again. - Don't retry
429s in a tight loop: every attempt counts towards the window.
Roadmap
Coming: limits per API key instead of per IP address, higher default limits for live keys, and RateLimit-Limit, RateLimit-Remaining, RateLimit-Reset and Retry-After headers on every response.
Base URL https://api.anyonein.co.uk/v1 is a placeholder until the production domain is confirmed. Found a mistake? Tell us.