Back to challenges
Challenge

Ayush Health DPI Triage API using Vercel AI SDK & Readme

Public health portals require compliant multilingual triage while verifying patient consent across Ayush health registries. Build a interactive health routing API using Vercel AI SDK and Readme to automate patient symptom triage and generate standardized consent-backed API docs.

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.

Develop a TypeScript patient triage service integrated with Ayush registry schemas using Vercel AI SDK and Readme API documentation tools.

Delivery guide

How work is evaluated

Evaluation

Evaluates health query classification correctness and OpenAPI schema compliance generated by the system.

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

check_abha_validation

Ensures requests without valid ABHA consent tokens are rejected

Binary check

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

Dimension 2

triage_accuracy

Correctness of department routing based on medical guidelines • target: 0.9 • 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 multi-provider streaming and tool calls using Vercel AI SDK

  • Auto-generate OpenAPI interactive specs using Readme SDK tooling

  • Parse and validate patient consent tokens against Ayush Digital Health standards

  • Support multilingual Hindi/English medical query routing

Resources and assets

Reference links and supporting material

Dataset notes

Multilingual Indian health query dataset annotated with ABDM consent metadata and Ayush diagnostic taxonomies.

How this agent runs

Evaluates health query classification correctness and OpenAPI schema compliance generated by the system.

Challenge input

JSON object with patient_symptoms, language, and abha_consent_token

AI SDK

Vercel's TypeScript SDK for building AI applications with tools and streaming.

Readme

Developer hub platform for generating and managing API documentation.

Evaluated output

JSON response with triage_category, specialization, and risk_urgency

Checks for
  • Ensures requests without valid ABHA consent tokens are rejected
  • Correctness of department routing based on medical guidelines • target: 0.9 • range: 0-1
Proof of success
  • Benchmark: Multilingual Medical Triage Benchmark
  • Triage Accuracy target: 0.9
  • 1 public reference case
Runtime evidence
  • JavaScript execution harness
  • javascript sandbox (unavailable on Versalist)
View technical recipe

Configured tools

Action Space
  • Readme · Required
  • Vercel AI SDK · Optional
  • LangWatch · Optional
Orchestration
  • Vercel AI SDK · Optional

Evaluation contract

  • check_abha_validation · Weight 1
  • triage_accuracy · Weight 1

Recipe state

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

Frequently Asked Questions about Ayush Health DPI Triage API using Vercel AI SDK & Readme