forecast-output-safety-validator

What problem does it solve?

This Skill provides a safety validator for meal forecast outputs in health AI applications, preventing unauthorized dosing advice, ensuring the integrity of structured data, and maintaining the educational nature of forecasts.

Core Features & Use Cases

• Safety Validation: Blocks dosing advice and checks structured data for compliance.
• Narrative Sanitization: Removes or replaces harmful content in narratives, preserving educational insights.
• Confidence Bound Checking: Ensures confidence levels align with evidence reliability.
• Use Case: For instance, before providing glucose impact predictions, this Skill would prevent dosing recommendations and verify the confidence of the prediction aligns with available data.

Quick Start

Use the forecast-output-safety-validator skill to validate a meal forecast output, ensuring it adheres to safety and educational guidelines.