Concepts
Monitored values are exact matches, using the same normalization as the screening API:
Plan limits: 25 monitored values per organization on Developer, 100 on Team, and per organization up to 3 webhook endpoints and 5 email addresses.
Add values one at a time, or paste a list of up to 100 of one type (one per line, with an optional label after a comma or in a second spreadsheet column). Each line is reported separately: added, already monitored, invalid, or over the plan limit, so a partly bad list still adds the good lines.
Webhook endpoints are public
https URLs owned by your organization. Each has its own signing secret (view and rotate it from the dashboard); the payload format and verification code are on the webhook events page.
Email addresses receive one message per event with the monitored value and label, the list, the listed value and party, the time, the event id, and a link to the official source record — the same facts the webhook payload carries. The address does not have to belong to a dashboard user, so a shared compliance inbox or a ticketing system’s intake address works; every message names the organization whose monitoring produced it. Nothing in an alert email is signed, so treat it as a notification to act on, and the webhook as the record to automate from.
Events fire per list: if a wallet is added to both the OFAC SDN list and the UK Sanctions List, you get two entity.listed events, each naming its list. Event types:
entity.listed— the value appeared on a list it wasn’t on beforeentity.delisted— the value was removed from a listentity.relisted— the value returned to a list it had been removed fromtest— sent by the dashboard’s test button
Delivery and retries
Events are delivered within a few minutes of ingestion. A webhook delivery is considered successful when your endpoint answers any 2xx status within 10 seconds; an email delivery when our mail provider accepts the message. Anything else is retried on a backoff schedule of roughly 1 minute, 5 minutes, 30 minutes, 2 hours, and 12 hours before the delivery is marked failed. Delivery is at-least-once and unordered: the same event can occasionally arrive twice, and two events may arrive out of order. For webhooks, deduplicate on the eventid (stable across retries) and use created_at to order; alert emails carry the same id.
After 5 consecutive failed deliveries (each having exhausted its retries), the destination is automatically disabled and the organization owner is emailed. Fix it and re-enable it from the dashboard — delivery resumes with new events from that point; events that occurred while it was disabled are not replayed.
An alert whose delivery failed can be redelivered from the dashboard’s Recent alerts list once the destination works again (organization owner only; the destination must be enabled). Redelivery attempts it immediately with the same event id, so your deduplication still applies; if that attempt fails, it goes back on the retry schedule above. Only deliveries that were attempted can be redelivered — events that occurred while a destination was disabled, or while monitoring was paused on the Free plan, were never queued for it.
The dashboard’s test button sends a test event to any destination: a signed POST to a webhook endpoint, or a sample alert email to an address, so you can confirm the wiring before anything real fires.
Verifying signatures
Every delivery carries anX-CompliAPI-Signature header signed with the endpoint’s secret. Always verify it before trusting a payload — see webhook events for the payload format and verification code in Python and Node.
Alerts vs. polling
The publicGET /api/v1/delisted feed is the polling counterpart: it lists recent removals across all lists, unauthenticated. Monitoring alerts are the push counterpart, scoped to exactly the entities you monitor.