Skip to main content
Back to challenges
Challenge

Municipal Debt Housing Recovery Underwriting Agent with Claude SDK

Municipal bond financing for disaster housing requires multi-factor wildfire risk and bond debt service evaluation. Build a Claude Agents SDK system with extended thinking to analyze bond prospectuses and rebuilding schedules, delivering full risk-adjusted municipal yield profiles.

Business OperationsHosted by Vera
Sign in free to start
Challenge brief

What you are building

The core problem, expected build, and operating context for this challenge.

Construct an autonomous underwriting assistant using Claude Agents SDK featuring extended thinking and tool use for debt coverage validation.

Delivery guide

How work is evaluated

Evaluation

Evaluates accuracy of bond debt service coverage calculation under wildfire delay stress testing.

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

dscr_stress_test

Stressed DSCR must be strictly less than base DSCR when delay > 0

Binary check

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

Dimension 2

underwriting_precision

Accuracy of debt service computations against benchmark bond models • target: 0.96 • 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

  • Deploy a Claude Agents SDK agent utilizing extended thinking capabilities.

  • Parse municipal bond offering prospectuses and disaster rebuilding risk disclosures.

  • Model debt service coverage ratios (DSCR) under insurance payout lag scenarios.

  • Synthesize risk-adjusted yield profiles with step-by-step reasoning logs.

Resources and assets

Reference links and supporting material

Dataset notes

Synthetic municipal bond prospectus disclosures, disaster insurance payout schedules, and home rebuilding timelines.

Challenge setup

Evaluates accuracy of bond debt service coverage calculation under wildfire delay stress testing.

Challenge input

JSON containing bond_issuance_amount, coupon_rate, annual_repayment_cap, insurance_delay_months

Claude Agents SDK

Framework assignment for Anthropic agent development.

Evaluated output

JSON with base_dscr, stressed_dscr, default_risk_flag boolean

Evaluation checks
  • Stressed DSCR must be strictly less than base DSCR when delay > 0
  • Accuracy of debt service computations against benchmark bond models • target: 0.96 • range: 0-1
Configured evaluation evidence
  • Benchmark: Claude-Finance-Bench
  • Underwriting Precision target: 0.96
  • 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
  • Aim · Required
  • Weights & Biases · Optional
  • Ellipsis · Optional

Evaluation contract

  • dscr_stress_test · Weight 1
  • underwriting_precision · Weight 1

Metrics

  • underwriting_precision · Target 0.96

Benchmarks

  • Claude-Finance-Bench
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON containing bond_issuance_amount, coupon_rate, annual_repayment_cap, insurance_delay_months

    Output: JSON with base_dscr, stressed_dscr, default_risk_flag boolean

    Execution harness: Python

Evaluation tools

  • Claude Agents SDK · Agent Framework

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Municipal Debt Housing Recovery Underwriting Agent with Claude SDK