Skip to main content
Back to challenges
Challenge

Claude Agents SDK Urban Land Tender Bidding Assistant

Urban development authorities release prime land tender sites with complex zoning, plot ratio, and usage guidelines. Construct an autonomous land tender evaluation agent using Claude Agents SDK to calculate gross floor area, identify land use constraints, and estimate tender bids.

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.

Build an urban planning and tender evaluation agent using Claude Agents SDK to analyze Singapore URA land sales announcements.

Delivery guide

How work is evaluated

Evaluation

Evaluates land tender metrics calculations and bid valuation accuracy.

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

gfa_math_validation

Ensures max_gfa_sqm equals site_area_sqm * gross_plot_ratio exactly.

Binary check

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

Dimension 2

bid_valuation_error

Percentage error between calculated residual land value and ground truth valuation. • target: 0.01 • range: 0-0.1

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

  • Implement Claude Agents SDK with custom tool bindings for urban planning geometry

  • Parse site area, plot ratio, and allowable gross floor area (GFA) from URA releases

  • Enforce zoning overlay rules and transit-oriented development (TOD) requirements

  • Generate automated tender bid recommendation models based on land rate benchmarks

Resources and assets

Reference links and supporting material

Dataset notes

Official media releases and land site specifications published by Singapore Urban Redevelopment Authority (URA).

Challenge setup

Evaluates land tender metrics calculations and bid valuation accuracy.

Challenge input

JSON with site_name, site_area_sqm, gross_plot_ratio, benchmark_psm_gfa

Claude Agents SDK

Provides Anthropic agent tool abstractions and extended thinking capabilities.

Evaluated output

JSON with max_gfa_sqm, estimated_tender_bid_sgd, zoning_status

Evaluation checks
  • Ensures max_gfa_sqm equals site_area_sqm * gross_plot_ratio exactly.
  • Percentage error between calculated residual land value and ground truth valuation. • target: 0.01 • range:...
Configured evaluation evidence
  • Benchmark: UrbanPlanningTenderBench
  • Bid Valuation Error target: 0.01
  • 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
  • RAI · Required
  • Resemble AI · Optional
  • Larridin · Optional

Evaluation contract

  • gfa_math_validation · Weight 1
  • bid_valuation_error · Weight 1

Metrics

  • bid_valuation_error · Target 0.01

Benchmarks

  • UrbanPlanningTenderBench
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON with site_name, site_area_sqm, gross_plot_ratio, benchmark_psm_gfa

    Output: JSON with max_gfa_sqm, estimated_tender_bid_sgd, zoning_status

    Execution harness: Python

Evaluation tools

  • Claude Agents SDK · Agent Framework

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Claude Agents SDK Urban Land Tender Bidding Assistant