Status & reliability
How the platform is operated, what to expect during incidents, and how to build a resilient integration on top of it.
Idempotency
All three v1 endpoints — /verify, /search, and /business/{id} — are read-only. Calling them multiple times with the same inputs is safe and produces the same result (subject to registry data updates). Network-level retries on timeout are safe.
Recommended client posture
- Use a request timeout of 10 seconds.
- Retry on
5xxand network errors with exponential backoff (1s → 2s → 4s → 8s, max 4 attempts). - Do not retry on
4xxwithout fixing the request. - Log
trustregistry_idalongside your own KYB record — it is stable and lets you correlate audits later.
Rate limits
There is no enforced rate limit on the API today. We operate the platform under a fair-use posture and reach out to accounts that produce sustained traffic spikes before taking action. Enterprise contracts include SLA-backed throughput targets — see sales@trustregistry.dev.
Incident communication
Active incidents and degraded-service notices are posted to the status page. For incidents affecting a specific account, our team will reach out directly to the account owner email on file.