What problem does it solve? New users of a health coaching bot need a structured first session that captures their profile (weight, height, sex, age, waist, goals, language, dietary restrictions) and turns it into a personalized plan, without losing track of missing inputs or inventing answers. ## Core Features & Use Cases - Loop-based intake checklist: Tracks each required input, validates values against sanity ranges (height 120–220 cm, weight 30–400 kg, age 13–100, waist 40–200 cm), and re-asks only for missing or out-of-range items. - Derived plan generation: Computes waist-to-height ratio (WHtR), selects a fat-loss or recomposition path by WHtR band, and sets weekly rate and protein targets scaled to the individual body. - Persistence handoff: Delegates database creation, profile.yaml writing, memory seeding, and reminder registration to the food-tracker first-run guard, then re-verifies readiness. - Use Case: A new tenant messages the bot for the first time; the skill greets them, collects their stats across multiple replies, derives a WHtR-based plan, and confirms setup once the guard prints READY. ## Quick Start Start the onboarding session for a new tenant who has no profile yet and walk them through the intake questions.
