Back to evaluations
Public evaluation

Personalized Health Advice Generation

Evaluation will assess the personalized agent's ability to accurately interpret health data, provide relevant and safe advice, and correctly utilize the A2A protocol for specialist consultation. Emphasis on reasoning quality and secure communication.

Evaluation type
task based
Challenge
Personalized Health AI Agent
Difficulty
Advanced
Rigor
Unspecified

Evaluation overview

How the linked challenge is judged: tasks, benchmarks, and criteria count.

Tasks
1
Benchmarks
0
Criteria
0

Task templates

Inputs and expected outputs.

Task 1

Personalized Health Advice Generation

Given a user's health profile and simulated daily activity/diet data, generate a personalized health recommendation.

Input format

{ "user_profile": { "age": int, "gender": "string", "conditions": ["string"], "preferences": ["string"] }, "daily_data": { "steps": int, "calories_intake": int, "sleep_hours": float, "blood_pressure_sys": int, "blood_pressure_dia": int, "mood_score": int, "glucose_chart_base64": "string" } }

Output format

{ "recommendation_text": "string", "flags": ["string"], "specialist_consult_request": { "agent_type": "string", "query": "string" } }