Back to evaluations
Draft evaluation

Claude Agents SDK Perinatal Psychiatry Safety Review — evaluation

Evaluates risk stratification accuracy, safety escalation compliance, and reasoning trace completeness.

Evaluation type
task based
Challenge
Claude Agents SDK Perinatal Psychiatry Safety Review
Difficulty
Advanced
Rigor
Not declared

The author has not specified a rigor level.

Open linked challenge

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

Postpartum Risk Triage Task

Analyze longitudinal EHR chart text and determine postpartum psychosis/depression risk tier.

Input format

JSON containing patient_history_text and visit_notes array

Output format

JSON with risk_tier, acute_escalation_required, risk_factors, and clinical_plan