Skip to main content
Loading navigation
Back to challenges

Build a Legal Entity Identifier Graph Resolver using Google ADK

Unassigned or orphaned Legal Entity Identifier (LEI) records create compliance vulnerabilities in international trade settlements. Build a legal entity verification agent using Google ADK with Gemini 1.5 Pro to traverse orphaned LEI portfolios and resolve corporate ownership chains. Achieve 94% accuracy in identifying ultimate parent companies across fragmented records.

CybersecurityHosted by Vera
Sign in free to start

What you are building

Develop an entity graph verification workflow with Google ADK to map orphaned LEI records back to ultimate parent entities.

How work is evaluated

Evaluation

Assess the agent's ability to accurately trace orphaned corporate LEI records to their true ultimate parent.

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

Accurate Parent LEI

Assert the output upe_lei matches ground truth ultimate parent.

Binary check

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

Dimension 2

UPE Identification Accuracy

Accuracy in mapping orphaned entities to correct UPE LEI. • target: 0.94 • 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

  • Implement graph traversal tools using Google Agent Development Kit (ADK)

  • Leverage Gemini 1.5 Pro long-context window to analyze multi-page corporate filing histories

  • Resolve orphaned LEI records to verified Ultimate Parent Entities (UPE)

  • Automate reporting to the Global Legal Entity Identifier Foundation (GLEIF)

Resources

Dataset notes

Synthetic dataset of 1,000 corporate hierarchy trees containing orphan and stale LEI entity nodes.

Challenge setup

Assess the agent's ability to accurately trace orphaned corporate LEI records to their true ultimate parent.

Challenge input

JSON containing orphan_lei, firm_name, jurisdiction, and historical_filings text.

Google ADK

Provides robust tools integration and Gemini model orchestration capabilities.

Evaluated output

JSON with upe_lei, upe_name, confidence, and chain_path.

Evaluation checks
  • Assert the output upe_lei matches ground truth ultimate parent.
  • Accuracy in mapping orphaned entities to correct UPE LEI. • target: 0.94 • range: 0-1
Configured evaluation evidence
  • Benchmark: LEIResolutionBench
  • UPE Identification Accuracy target: 0.94
  • 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
  • Google · Required
  • Google ADK · Optional
  • Google Jules · Optional
Orchestration
  • Google Jules · Optional

Evaluation contract

  • Accurate Parent LEI · Weight 1
  • UPE Identification Accuracy · Weight 1

Metrics

  • UPE Identification Accuracy · Target 0.94

Benchmarks

  • LEIResolutionBench
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON containing orphan_lei, firm_name, jurisdiction, and historical_filings text.

    Output: JSON with upe_lei, upe_name, confidence, and chain_path.

    Execution harness: Python

Evaluation tools

  • Google ADK · Agent Development Kit

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Build a Legal Entity Identifier Graph Resolver using Google ADK