Browse Tasks

Select category

Select category

Select category

Select category

Select Task

Select Task

Select Task

Select Task

Select Tasks

Select Industry

Customer Onboarding Pre-triage

Investment Portfolio Analysis

Compliance and Regulatory Check

Fraud Detection

Transaction Categorisation

Financial Product Recommendation

News Article Summary

Fraud Analyst

Fraud Detection

This agentic task detects fraudulent financial transactions - essential for compliance and monitoring agents. The task configuration outlines how to:

  • Analyse transaction patterns and behavioural indicators

  • Evaluate historical account activity alignment

  • Detect anomalous transaction characteristics

  • Calculate risk scores for suspicious activities

  • Generate immediate response recommendations


As a modular agentic task, it integrates with transaction processing and security systems, enabling AI agents to execute real-time fraud detection within existing monitoring workflows. The task specification includes frameworks for pattern analysis, risk assessment, and alert generation, adaptable across banking, payment processing, and e-commerce platforms.

Example outputs:

{
  "fraud_risk_score": "0.85",
  "risk_level": "High",
  "suspicious_factors": [
    "Transaction location inconsistent with account history",
    "Unusually large transaction amount",
    "Multiple high-value transactions in short timeframe",
    "Transaction occurred outside typical account usage hours"
  ],
  "recommended_action": "Block Transaction",
  "explanation": "The transaction shows multiple high-risk factors including an atypical location, unusually large amount, and occurrence outside normal hours. This pattern significantly deviates from the account's historical behavior.",
  "additional_verification_needed": [
    "Contact account holder for confirmation",
    "Request additional identity verification",
    "Verify recent travel plans with account holder"