Skills Compliance

Fraud and scam signals in support conversations

Support hears about fraud before fraud systems do. Transaction monitoring sees a payment that fits a pattern; a support agent hears a customer say someone called them claiming to be from the bank. The second is earlier, richer, and usually not analysed.

ComplianceFinancial servicesAnalysisAny helpdeskRead-only
Installnpx rulebase-skills install cx-fraud-and-scam-signal

When to use it

Reach for this when someone says any of these — they are the phrases the skill itself triggers on:

  • are customers reporting scams
  • new scam pattern targeting our customers
  • did we spot the fraud signals
  • customer was coached by someone on the phone

How it works

The method, in the order the skill runs it. The full procedure — tables, worked examples and the edge cases — is in the skill itself.

  1. Signal categories

    Scam-in-progress, the highest-value and most time-critical. Customers describe the mechanics without recognising them: Unusual urgency tied to a threat, An investment opportunity with a contact who has been messaging them, A romance or long-relationship framing around a payment.

  2. Why support text beats transaction data here

    The payment looks legitimate. The customer authorised it, from their own device, in a normal pattern. The only evidence that it was a scam is in what they said — and that evidence is frequently in a support conversation days before the loss, or in the same conversation as the payment.

  3. Handling assurance

    For conversations with a signal present: Was it recognised? Or did the agent process the request and move on?, Was the required intervention attempted, Was it routed to the specialist team, and how quickly?.

  4. Feeding it back

    The durable output is not a list of cases — it is what changes: Product friction findings.

Related skills

Free and open source, and vendor-neutral — it reads the conversations from whichever helpdesk you already run. Browse all 149 skills · connect your helpdesk over MCP · source on GitHub

Review every conversation. Act on what it finds.

AI for customer operations, built for financial services. Specialist agents chase every issue to resolution and every stalled customer to activation.

Rulebase dashboard