Skip to main content
Back to challenges
Challenge

Build a MAS Compliance Audit Swarm with CrewAI and /dev/agents

Multi-asset fund managers struggle to maintain continuous regulatory compliance across dynamic Singapore fund structures. Deploy a collaborative multi-agent CrewAI swarm integrated with /dev/agents to audit asset allocations against MAS guidelines and automatically generate structured rebalancing instructions.

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.

Deploy a collaborative CrewAI agent team using /dev/agents to analyze asset management portfolios for MAS regulatory compliance and generate structured rebalancing instructions.

Delivery guide

How work is evaluated

Evaluation

Evaluates CrewAI swarm accuracy in identifying portfolio concentration breaches and emitting correct rebalancing actions.

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

violation_detection_pass

Ensures all asset threshold breaches are detected

Binary check

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

Dimension 2

rebalance_accuracy

Percentage of rebalance orders that resolve limits without creating new breaches • target: 0.95 • 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 role-based AI agents using CrewAI framework

  • Integrate /dev/agents runtime environment for reliable tool execution

  • Parse portfolio allocations against MAS statutory limits

  • Generate verifiable rebalancing order tickets with full audit trails

Resources and assets

Reference links and supporting material

Dataset notes

Synthetic multi-asset fund holdings data and MAS investment limit parameter sets.

Challenge setup

Evaluates CrewAI swarm accuracy in identifying portfolio concentration breaches and emitting correct rebalancing actions.

Challenge input

JSON object with portfolio holdings and MAS regulatory constraints

CrewAI

Supports role-based agent collaboration and delegation

/dev/agents

Provides tool execution environment for quantitative portfolio tools

Evaluated output

JSON object with compliance_status, violations array, and rebalance_orders array

Evaluation checks
  • Ensures all asset threshold breaches are detected
  • Percentage of rebalance orders that resolve limits without creating new breaches • target: 0.95 • range: 0-1
Configured evaluation evidence
  • Benchmark: PortfolioAudit-2026
  • Rebalance Accuracy 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
  • CrewAI · Required
  • /dev/agents · Optional
  • Traceloop · Optional

Evaluation contract

  • violation_detection_pass · Weight 1
  • rebalance_accuracy · Weight 1

Metrics

  • rebalance_accuracy · Target 0.95

Benchmarks

  • PortfolioAudit-2026
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON object with portfolio holdings and MAS regulatory constraints

    Output: JSON object with compliance_status, violations array, and rebalance_orders array

    Execution harness: Python

Evaluation tools

  • CrewAI · Agent Framework
  • /dev/agents · Agent System Runtime

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Build a MAS Compliance Audit Swarm with CrewAI and /dev/agents