Back to challenges
Challenge

Multi-Agent VASP Licensing Risk Auditing with AutoGen and Oracle OCI GenAI

Regulators and compliance teams require robust multi-agent dialogue systems when auditing Virtual Asset Service Providers (VASPs). Build an AutoGen multi-agent system powered by Oracle OCI Generative AI to evaluate VASP operational compliance, custody security, and anti-money laundering controls.

Workflow AutomationHosted by Vera
Sign in free to start
Challenge brief

What you are building

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

Orchestrate an AutoGen multi-agent system backed by Oracle OCI Generative AI to perform automated risk assessments for virtual asset service providers.

Delivery guide

How work is evaluated

Evaluation

Evaluates multi-agent conversation outcomes in classifying VASP risk profiles and wallet custody parameters.

Datasets

Shared data for this challenge

Review public datasets and any private uploads tied to your build.

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

travel_rule_rejection

Verifies application is rejected or flagged conditional if Travel Rule is missing

Binary check

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

Dimension 2

audit_thoroughness

Percentage of key compliance risks accurately identified in summary • target: 0.9 • 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

  • Orchestrate conversational multi-agent systems using AutoGen framework

  • Connect AutoGen agents to Oracle OCI Generative AI inference endpoints

  • Evaluate VASP licensing applications against MAS Payment Services Act rules

  • Incorporate human-in-the-loop sign-off steps for risk exception approvals

Resources and assets

Reference links and supporting material

Dataset notes

Synthetic licensing filings for Singapore Virtual Asset Service Providers.

How this agent runs

Evaluates multi-agent conversation outcomes in classifying VASP risk profiles and wallet custody parameters.

Challenge input

JSON object with vasp_name, cold_wallet_ratio, travel_rule_solution, reserve_audit_frequency

AutoGen

Enables multi-turn conversational debate between specialized agents

Oracle OCI Generative AI

Provides enterprise hosted LLM backends

Evaluated output

JSON object with licensing_recommendation, key_risks, and required_remediations

Checks for
  • Verifies application is rejected or flagged conditional if Travel Rule is missing
  • Percentage of key compliance risks accurately identified in summary • target: 0.9 • range: 0-1
Proof of success
  • Benchmark: VASP-Audit-Bench
  • Audit Thoroughness target: 0.9
  • 1 public reference case
Runtime evidence
  • Python execution harness
  • Python sandbox (unavailable on Versalist)
View technical recipe

Configured tools

Action Space
  • AutoGen · Required
  • Oracle OCI Generative AI · Optional
  • Dora · Optional

Evaluation contract

  • travel_rule_rejection · Weight 1
  • audit_thoroughness · Weight 1

Recipe state

This is a preview. The configuration can change before the evaluation recipe is locked.

Frequently Asked Questions about Multi-Agent VASP Licensing Risk Auditing with AutoGen and Oracle OCI GenAI