Skip to main content
Back to challenges
Challenge

Build a Small-Bank Liquidity Stress Testing & Failure Prediction Agent with AutoGen

Small community bank failures threaten regional financial stability during liquidity shocks. Build a multi-agent AutoGen stress-testing system where risk, liquidity, and macroeconomic agents simulate deposit flight scenarios and predict bank failure thresholds with 94% forecast precision.

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.

Utilize Microsoft AutoGen multi-agent simulation to test community bank balance sheets under rapid deposit outflow stress scenarios.

Delivery guide

How work is evaluated

Evaluation

Evaluates failure prediction accuracy and capital depletion timeline precision.

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

insolvency_trigger_test

Must correctly identify insolvency trigger when Tier 1 capital goes negative.

Binary check

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

Dimension 2

forecast_precision

Precision of bank failure prediction across test cases. • target: 0.94 • range: 0.85-1

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

  • Build multi-agent financial stress simulation environments using AutoGen

  • Simulate unrealized bond losses and uninsured deposit runs

  • Predict regulatory intervention points (FDIC receivership triggers)

  • Generate automated capital restoration recommendations for bank boards

Resources and assets

Reference links and supporting material

Dataset notes

Dataset of 100 small community bank balance sheets with labeled liquidity stress outcomes.

Challenge setup

Evaluates failure prediction accuracy and capital depletion timeline precision.

Challenge input

JSON balance sheet payload including uninsured deposit ratio and HTM securities losses

AutoGen

Allows complex role-playing simulation between financial risk personas.

Evaluated output

JSON failure probability, days to insolvency, and key vulnerability factor

Evaluation checks
  • Must correctly identify insolvency trigger when Tier 1 capital goes negative.
  • Precision of bank failure prediction across test cases. • target: 0.94 • range: 0.85-1
Configured evaluation evidence
  • Benchmark: Bank Liquidity Stress Suite
  • Forecast Precision target: 0.94
  • 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
  • AutoGen · Required
  • Microsoft AutoGen · Optional

Evaluation contract

  • insolvency_trigger_test · Weight 1
  • forecast_precision · Weight 1

Metrics

  • forecast_precision · Target 0.94

Benchmarks

  • Bank Liquidity Stress Suite
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON balance sheet payload including uninsured deposit ratio and HTM securities losses

    Output: JSON failure probability, days to insolvency, and key vulnerability factor

    Execution harness: Python

Evaluation tools

  • AutoGen · Multi-Agent Simulation Engine

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Build a Small-Bank Liquidity Stress Testing & Failure Prediction Agent with AutoGen