Skip to main content
Back to challenges
Challenge

Build a Cross-Border Sanction Screening Agent with Google ADK and Google Jules

Post-sanction regulatory transitions require immediate re-verification of transaction entities across jurisdiction boundaries. Build a Google ADK agent integrated with Google Jules to analyze historical cross-border clearing logs, evaluate lifted sanctions against OFAC updates, and route compliant transactions safely. Achieve deterministic verification speed with zero compliance leakage across modified sanction lists.

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.

Construct an enterprise sanction screening pipeline using Google ADK with Gemini 1.5 Pro and Google Jules AI code orchestration for real-time OFAC update verification.

Delivery guide

How work is evaluated

Evaluation

Tests the agent's accuracy in re-screening cross-border transactions following modified jurisdictional sanction regimes.

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

sanction_status_accuracy

Verifies correct legal permit status identification

Binary check

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

Dimension 2

compliance_accuracy

Proportion of accurately routed cross-border transactions • target: 0.98 • range: 0-1

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

  • Develop a sanction resolution agent using Google ADK and Gemini models

  • Use Google Jules agentic automation to dynamically update local compliance rules

  • Evaluate transaction counterparties against dynamically updated sanction registries

  • Ensure regulatory auditability with explainable rule-match telemetry

Resources and assets

Reference links and supporting material

Dataset notes

Dataset of 300 cross-border ISO 20022 payment messages involving post-sanction transition entities.

Challenge setup

Tests the agent's accuracy in re-screening cross-border transactions following modified jurisdictional sanction regimes.

Challenge input

JSON containing transaction details, counterparty bank LEI, and country code

Google ADK

Provides Vertex AI native Gemini integration and function calling capabilities.

Google Jules

Automates rule base updates and codebase synchronization.

Evaluated output

JSON with sanction_status, rule_version, and approval_flag

Evaluation checks
  • Verifies correct legal permit status identification
  • Proportion of accurately routed cross-border transactions • target: 0.98 • range: 0-1
Configured evaluation evidence
  • Benchmark: OFAC Regulatory Compliance Suite
  • Compliance Accuracy target: 0.98
  • 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
  • Google Jules · Required
  • Google · Optional
  • Google ADK · Optional
Orchestration
  • Google Jules · Required

Evaluation contract

  • sanction_status_accuracy · Weight 1
  • compliance_accuracy · Weight 1

Metrics

  • compliance_accuracy · Target 0.98

Benchmarks

  • OFAC Regulatory Compliance Suite
  • security

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON containing transaction details, counterparty bank LEI, and country code

    Output: JSON with sanction_status, rule_version, and approval_flag

    Execution harness: Python

Evaluation tools

  • Google ADK · Agent Framework
  • Google Jules · AI Engineering Orchestration

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Build a Cross-Border Sanction Screening Agent with Google ADK and Google Jules