Skip to main content
Back to challenges
Challenge

Build an OpenAI Agents SDK Aadhaar Identity Verification Triage System

High-volume citizen onboarding services need rapid verification of OCR-extracted identity documents against Aadhaar metadata formatting standard constraints. Build an OpenAI Agents SDK identity verification agent that validates document tokens and flags fraudulent attempts with 98% accuracy.

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 identity verification agent using OpenAI Agents SDK to validate Aadhaar metadata schema structures and anomaly indicators.

Delivery guide

How work is evaluated

Evaluation

Evaluates checksum validity verification and PII masking completeness.

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

pii_masking_check

Confirms identity numbers are properly masked

Binary check

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

Dimension 2

verification_accuracy

Accuracy of identity document authenticity decisions • 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

  • Implement OpenAI Agents SDK agents with custom tool functions for Verhoeff checksum validation

  • Detect structural anomalies in OCR extracted identity payloads

  • Enforce privacy protection masking on citizen identity numbers

  • Generate encrypted KYC audit payloads compliant with UIDAI standards

Resources and assets

Reference links and supporting material

Dataset notes

Synthetic identity document OCR text Extractions and metadata records.

Challenge setup

Evaluates checksum validity verification and PII masking completeness.

Challenge input

JSON identity OCR extraction object

OpenAI Agents SDK

Agent loop execution and function calling

Evaluated output

JSON audit result with checksum_valid and masked_id

Evaluation checks
  • Confirms identity numbers are properly masked
  • Accuracy of identity document authenticity decisions • target: 0.98 • range: 0-1
Configured evaluation evidence
  • Benchmark: Identity Document Parsing Benchmark
  • Verification Accuracy target: 0.98
  • 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
  • OpenAI · Required
  • RAI · Optional
  • LangWatch · Optional

Evaluation contract

  • pii_masking_check · Weight 1
  • verification_accuracy · Weight 1

Metrics

  • verification_accuracy · Target 0.98

Benchmarks

  • Identity Document Parsing Benchmark
  • security

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON identity OCR extraction object

    Output: JSON audit result with checksum_valid and masked_id

    Execution harness: Python

Evaluation tools

  • OpenAI Agents SDK · Agent Framework

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Build an OpenAI Agents SDK Aadhaar Identity Verification Triage System