What is the 'Data Separation Principle' mentioned in this skill?

It is a testing methodology that keeps historical records in the database while placing current symptoms in the user prompt, ensuring the AI uses diagnostic reasoning rather than simple data retrieval.

What are the success criteria for passing these tests?

Individual test cases require an 85% pass rate across success criteria, and the suite is only considered successful if there are zero critical safety failures.

What is the primary purpose of the Saudi Patient Testing skill?

It provides a structured guide and automated suite to test the diagnostic accuracy and clinical safety of healthcare AI platforms using specific Saudi patient scenarios.

Which safety protocols does this skill validate?

It specifically tests for penicillin allergy awareness, antibiotic stewardship (avoiding over-prescription for viral illnesses), and the ability to identify urgent surgical emergencies.

Saudi Telehealth AI Testing

Name: Saudi Telehealth AI Testing
Author: adaptationio

byadaptationio

セキュリティとテスト

Validates diagnostic accuracy and clinical safety for AI-driven telehealth consultations involving Saudi patient profiles.

概要

This skill provides a comprehensive testing framework for evaluating AI diagnostic capabilities within the Saudi healthcare context, specifically for the Dr. Sophia AI platform. It utilizes five detailed patient personas to test complex clinical scenarios including chronic disease management, antibiotic stewardship, and urgent surgical referrals. By enforcing a strict Data Separation Principle—where historical data is isolated from current symptoms—it ensures the AI performs genuine clinical reasoning rather than simple information retrieval, maintaining an 85% success threshold for critical safety protocols like allergy verification.

主な機能

Automated test execution scripts for validating AI diagnostic accuracy
0 GitHub stars
Critical safety checks for penicillin allergies and emergency referral flagging
Enforcement of the Data Separation Principle to evaluate actual clinical reasoning
Pre-defined success metrics and weighted criteria for clinical decision-making
Standardized test cases for 5 diverse Saudi patient profiles with unique medical histories

ユースケース

Validating clinical safety protocols and allergy-checking mechanisms in telehealth bots
Benchmarking AI diagnostic accuracy for Saudi-specific medical consultations
Automating regression testing for AI healthcare models and clinical decision support systems

概要

主な機能

Automated test execution scripts for validating AI diagnostic accuracy
0 GitHub stars
Critical safety checks for penicillin allergies and emergency referral flagging
Enforcement of the Data Separation Principle to evaluate actual clinical reasoning
Pre-defined success metrics and weighted criteria for clinical decision-making
Standardized test cases for 5 diverse Saudi patient profiles with unique medical histories

ユースケース

Validating clinical safety protocols and allergy-checking mechanisms in telehealth bots
Benchmarking AI diagnostic accuracy for Saudi-specific medical consultations
Automating regression testing for AI healthcare models and clinical decision support systems