Skip to main content
Back to challenges
Challenge

Build a Synthetic Identity Fraud Detector with Claude Agents SDK

Synthetic identity fraud combines fake and real PII to bypass traditional credit checks. Build a Claude Agents SDK multi-step investigation agent that cross-references credit bureau records, utility bills, and social security traces to catch synthetic identities with 95% precision.

CybersecurityHosted by Vera
Sign in free to start
Challenge brief

What you are building

The core problem, expected build, and operating context for this challenge.

Utilize Claude Agents SDK extended context reasoning to perform multi-source entity validation for synthetic identity fraud detection.

Delivery guide

How work is evaluated

Evaluation

Evaluates synthetic identity detection precision and evidence generation accuracy.

Datasets

Challenge datasets

Public datasets, plus the private datasets you have access to.

Evaluation rubric

How submissions are scored

These dimensions define what the evaluator checks and which criteria separate a passable run from a strong one.

Dimensions
2 scoring checks
Binary
2 pass or fail dimensions
Ordinal
0 scaled dimensions
Dimension 1

ssn_mismatch_test

Must flag SSN issuance date mismatch relative to applicant age.

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Dimension 2

precision

Synthetic identity detection precision score. • target: 0.95 • range: 0.88-1

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Learning goals

What you should walk away with

  • Build multi-step identity verification agents with Claude Agents SDK

  • Perform cross-database correlation between SSN issuance dates and credit histories

  • Detect synthetic identity footprints across digital bank account applications

  • Generate detailed fraud evidence packages for bank loss prevention teams

Resources and assets

Reference links and supporting material

Dataset notes

Dataset of 1,000 synthetic identity profiles and legitimate customer profiles with credit bureau trace histories.

Challenge setup

Evaluates synthetic identity detection precision and evidence generation accuracy.

Challenge input

JSON customer application record with SSN, credit history, and phone tenure

Claude Agents SDK

Excels at nuanced multi-source factual verification tasks.

Evaluated output

JSON verdict, synthetic probability score, and primary evidence points

Evaluation checks
  • Must flag SSN issuance date mismatch relative to applicant age.
  • Synthetic identity detection precision score. • target: 0.95 • range: 0.88-1
Configured evaluation evidence
  • Benchmark: Synthetic Identity Suite 2026
  • Precision target: 0.95
  • 1 public reference case
Configured runtime
  • Python execution harness
  • Python sandbox (unavailable on Versalist)

Summary shows the first task, up to three tools, and up to three evidence items. Open the technical recipe to view all configured items.

View technical recipe

Configured tools

Action Space
  • Claude Code · Required
Observation
  • Claude Code · Required
Policy Serving
  • Claude 4.1 Opus · Optional
  • Claude 4 Sonnet · Optional

Evaluation contract

  • ssn_mismatch_test · Weight 1
  • precision · Weight 1

Metrics

  • precision · Target 0.95

Benchmarks

  • Synthetic Identity Suite 2026
  • security

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON customer application record with SSN, credit history, and phone tenure

    Output: JSON verdict, synthetic probability score, and primary evidence points

    Execution harness: Python

Evaluation tools

  • Claude Agents SDK · Reasoning Agent Framework

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Build a Synthetic Identity Fraud Detector with Claude Agents SDK