Skip to main content
Loading navigation
Back to challenges

Cross-Border Trade & Customs Clearance Orchestrator with CrewAI

Cross-border maritime trade documents between Singapore and regional ports require multi-party validation across customs, phytosanitary, and bill-of-lading forms. Build a collaborative multi-agent workforce using CrewAI to automate customs review. Reduce manual review processing time and achieve reliable exception routing.

Workflow AutomationHosted by Vera
Sign in free to start

What you are building

Build a collaborative multi-agent trade clearance automation workforce using CrewAI.

How work is evaluated

Evaluation

Evaluates document anomaly detection rate and multi-agent coordination correctness.

Datasets

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

How submissions are scored

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

hs_code_match_check

Ensures HS code evaluation aligns with declared tariff tables.

Binary check

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

Dimension 2

crew_task_completion_rate

Ratio of tasks successfully resolved without unhandled exceptions. • target: 1 • 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

  • Define CrewAI agents with dedicated roles: Customs Specialist, Fraud Examiner, Logistics Coordinator

  • Establish sequential and hierarchical task execution crews for multi-document parsing

  • Automate cross-matching between Bills of Lading, HS Code declarations, and packing slips

  • Implement human handoff triggers when customs discrepancies exceed severity bounds

Resources

Dataset notes

Synthetic cross-border shipping document manifest dataset including declared goods, invoice values, and HS codes.

Challenge setup

Evaluates document anomaly detection rate and multi-agent coordination correctness.

Challenge input

JSON customs packet containing declarations and bill of lading

CrewAI

Assigned framework for multi-agent delegation.

Evaluated output

JSON clearance status and agent logs

Evaluation checks
  • Ensures HS code evaluation aligns with declared tariff tables.
  • Ratio of tasks successfully resolved without unhandled exceptions. • target: 1 • range: 0-1
Configured evaluation evidence
  • Benchmark: TradeClear-MultiAgent-Bench
  • Crew Task Completion Rate target: 100%
  • 1 public reference case
Configured runtime
  • Python execution harness
  • Python sandbox (unavailable on Versalist)

Shows the first task, up to three tools, and up to three evidence items.

View technical recipe

Configured tools

Action Space
  • CrewAI · Required
  • crewAI · Optional
  • Decipher AI · Optional

Evaluation contract

  • hs_code_match_check · Weight 1
  • crew_task_completion_rate · Weight 1

Metrics

  • crew_task_completion_rate · Target 1

Benchmarks

  • TradeClear-MultiAgent-Bench
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON customs packet containing declarations and bill of lading

    Output: JSON clearance status and agent logs

    Execution harness: Python

Evaluation tools

  • CrewAI · Multi-Agent Orchestrator

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Cross-Border Trade & Customs Clearance Orchestrator with CrewAI