What problem does it solve? Users asking health questions often get single-answer guesses, fabricated statistics, or miss emergency red flags. This Skill enforces a safety-first workflow: red flag screening before any research, differential diagnosis instead of one answer, and citations from tier-1 medical sources like MedlinePlus, Mayo Clinic, NHS, Cochrane, and PubMed. ## Core Features & Use Cases - Symptom Triage: Screens for emergency red flags first, asks targeted clarifying questions, then produces a triage verdict with a ranked differential and escalation guidance. - Condition Research: Runs a multi-phase research plan across patient-facing and clinical sources, producing a structured evidence summary with evidence quality explicitly flagged. - Doctor Visit Preparation: Builds a printable prep card with symptoms, medications, history, and tailored questions using AHRQ visit-type question banks via scripts/prep_card.py. - Use Case: A user says "I've had mouth ulcers for years, what could it be?" The skill screens red flags, clarifies medication composition (e.g., Claritin vs. Claritin-D), builds an iterative differential across turns, and ends with a prep card of questions for their doctor. ## Quick Start Ask the agent to help you understand a symptom or prepare for a doctor visit, for example: "I have had fatigue and headaches for two months — screen for anything serious and help me prepare questions for my doctor."
