Back to challenges
Challenge

Build a Public Service Eligibility & Benefit Routing Agent with Claude Agents SDK

Citizens navigating family milestones need automated cross-agency service recommendations under Smart Nation initiatives like LifeSG. Build an autonomous agent using the Claude Agents SDK that ingests citizen profiles, evaluates eligibility rules, and generates personalized application workflows.

Workflow AutomationHosted 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 a citizen workflow agent using Claude Agents SDK to evaluate milestone benefits and automate multi-agency application payloads.

Delivery guide

How work is evaluated

Evaluation

Evaluates Claude Agents SDK agent accuracy in determining citizen benefit eligibility and producing valid multi-agency submission schemas.

Datasets

Shared data for this challenge

Review public datasets and any private uploads tied to your build.

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

baby_bonus_eligibility_pass

Ensures newborn profile triggers Baby Bonus eligibility inclusion

Binary check

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

Dimension 2

grant_calc_precision

Estimated grant total accuracy against government baseline matrix • target: 0.98 • 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

  • Build complex autonomous agents using the Claude Agents SDK

  • Leverage extended thinking capabilities for multi-step eligibility reasoning

  • Parse public scheme rules across health, housing, and childcare domains

  • Construct multi-agency API submission payloads for government services

Resources and assets

Reference links and supporting material

Dataset notes

Synthetic citizen demographic profiles and Singapore milestone grant benefit rules.

How this agent runs

Evaluates Claude Agents SDK agent accuracy in determining citizen benefit eligibility and producing valid multi-agency submission schemas.

Challenge input

JSON object with household_income_sgd, num_children, housing_type, newborn_date

Claude Agents SDK

Enables extended thinking and native tool loops with Claude

Evaluated output

JSON object with eligible_schemes array, total_estimated_grant_sgd, and action_plan steps

Checks for
  • Ensures newborn profile triggers Baby Bonus eligibility inclusion
  • Estimated grant total accuracy against government baseline matrix • target: 0.98 • range: 0-1
Proof of success
  • Benchmark: GovService-Bench-2026
  • Grant Calc Precision target: 0.98
  • 1 public reference case
Runtime evidence
  • Python execution harness
  • Python sandbox (unavailable on Versalist)
View technical recipe

Configured tools

Action Space
  • Zed · Required
  • BeeAgent · Optional
Observation
  • Langfuse · Optional
Reward / Eval
  • Langfuse · Optional

Evaluation contract

  • baby_bonus_eligibility_pass · Weight 1
  • grant_calc_precision · Weight 1

Recipe state

This is a preview. The configuration can change before the evaluation recipe is locked.

Frequently Asked Questions about Build a Public Service Eligibility & Benefit Routing Agent with Claude Agents SDK