Skip to main content
Back to challenges
Challenge

Cross-Border Financial Crime & Entity Resolution Agent Network with CrewAI

Financial intelligence units struggle to connect disparate offshore entity records during active cross-border fraud investigations. Build a collaborative multi-agent investigation network using CrewAI and WebAI to resolve obscure corporate entities, achieving 95% entity disambiguation accuracy across noisy records.

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.

Orchestrate specialized CrewAI role-playing agents paired with WebAI web extraction engines to trace complex corporate ownership structures and cross-border money laundering rings.

Delivery guide

How work is evaluated

Evaluation

Validates multi-agent CrewAI performance in linking hidden shell company networks and ultimate beneficial owners.

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

No False Disambiguation Merges

Ensure distinct legal entities with similar names are not falsely merged.

Binary check

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

Dimension 2

accuracy

Entity resolution match accuracy • target: 0.95 • range: 0.92-1

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

  • Design multi-agent CrewAI workflows with distinct investigator, auditor, and analyst roles

  • Integrate WebAI for web extraction and structural parsing of offshore corporate registries

  • Build dynamic entity disambiguation logic across multi-lingual regulatory disclosures

Resources and assets

Reference links and supporting material

Dataset notes

Multi-jurisdictional corporate registration dataset containing 5,000 ambiguous company filings with known synthetic UBO links.

Challenge setup

Validates multi-agent CrewAI performance in linking hidden shell company networks and ultimate beneficial owners.

Challenge input

JSON array of raw corporate registry records and individual profiles

CrewAI

Mandated agent framework for multi-role agent collaboration.

WebAI

Assigned tool for scraping web business registries.

Evaluated output

JSON entity graph with canonical entity IDs and confidence metrics

Evaluation checks
  • Ensure distinct legal entities with similar names are not falsely merged.
  • Entity resolution match accuracy • target: 0.95 • range: 0.92-1
Configured evaluation evidence
  • Benchmark: EntityRes-Bench
  • 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
  • crewAI · Optional
  • WebAI · Optional

Evaluation contract

  • No False Disambiguation Merges · Weight 1
  • accuracy · Weight 1

Metrics

  • accuracy · Target 0.95

Benchmarks

  • EntityRes-Bench
  • security

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON array of raw corporate registry records and individual profiles

    Output: JSON entity graph with canonical entity IDs and confidence metrics

    Execution harness: Python

Evaluation tools

  • CrewAI · Agent Framework
  • WebAI · Web Search & Extraction

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Cross-Border Financial Crime & Entity Resolution Agent Network with CrewAI