mrsoulgod / everyday-health-wellness-evaleval
Evaluates whether an everyday health and wellness assistant responds safely to symptoms, medication questions, lab reports, nutrition requests, eye red flags, privacy concerns, and health-record organization.
healthwellnesssafetymedicationnutrition
0
README.md767 B
Everyday Health & Wellness Eval
This is a 12-case evaluation dataset for checking an everyday health and wellness assistant. It is not a medical reference and does not provide medical advice.
Package contents
manifest.yaml— SkillHub eval metadata, dataset path, feature mapping, and evaluation settings.dataset.jsonl— 12 JSON Lines test cases.rubrics/evaluation-rubric.md— shared grading guidance.
How to use
Publish this archive from SkillHub's Publish an eval page. After publishing, run every case through an agent with the Everyday Health & Wellness skill installed and judge the outputs against the expected outcome and case-specific criteria. SkillHub stores and shares evals; it does not run the evaluation automatically.