Skip to main content
Alerts are automatically generated by FraudEG’s risk engine when it detects sustained or elevated fraud patterns across your account — not just a single suspicious transaction, but correlated activity that suggests a broader attack. Examples include a velocity spike in failed payment attempts, a cluster of new devices exhibiting account-takeover signals, or a pattern consistent with card testing. Unlike individual transaction events, alerts aggregate signals across multiple users or transactions and require your team to investigate and close them. Use this endpoint to list, filter, and triage active alerts in your fraud operations workflow.

Endpoint

Query Parameters

string
default:"open"
Filter by alert lifecycle status. Accepted values:
  • open — Alert is active and requires attention
  • acknowledged — Alert has been seen and is under investigation
  • resolved — Alert has been closed with a resolution
string
Filter by severity level. Accepted values: low, medium, high, critical. Critical alerts indicate an active, high-volume attack pattern.
string
Filter by the type of fraud pattern detected. Common values include:
  • velocity_spike — Abnormal transaction volume in a short window
  • new_device_cluster — Multiple new devices with correlated risk signals
  • ato_pattern — Signals consistent with account takeover attempts
  • card_testing — Sequential low-value charges probing card validity
  • identity_abuse — Reuse or fabrication of identity attributes
string
Filter to alerts that include a specific user in their affected scope. Useful when investigating a compromised account.
string
ISO 8601 timestamp. Returns only alerts created at or after this time.
string
ISO 8601 timestamp. Returns only alerts created before or at this time.
integer
default:"20"
Number of results to return per page. Accepted range: 1100.
string
Pagination cursor. Pass the next_cursor value from a previous response to fetch the next page.

Response Fields

array
Ordered list of alert objects, sorted by created_at descending.
boolean
true if additional pages of results exist beyond this response.
string | null
Opaque cursor for fetching the next page. Pass this as the after query parameter. Returns null on the last page.

Managing Alert Status

Once you retrieve an alert, you can update its status using PATCH /v1/alerts/{id}. The request body accepts a JSON object.

Acknowledge an alert

Send this request to indicate that your team is actively investigating the alert.

Resolve an alert

Send this request to close the alert and record your resolution findings.

Examples

Response

Instead of polling this endpoint, subscribe to the alert.triggered webhook event to receive real-time notifications the moment FraudEG generates a new alert. Webhook delivery is typically within two seconds of alert creation, making it the preferred integration pattern for time-sensitive fraud operations.