PRD: Maintain (MVP)

Goals

Help users define and stay within a maintenance range.
Make fluctuation legible through low-friction logging and calm recommendations.
Help users observe how GLP-1 support lines up with maintenance outcomes without turning the app into a prescriptive medical tool.
Preserve enough personal history and life-stage context to make maintenance feel individualized rather than generic.

Comprehensive medical management or a full clinician workflow in the first MVP.
Recommending medication changes without clear user control and clinical guardrails.

User sets a target weight and acceptable maintenance range during onboarding, defaulting to a 5 lb band but allowing custom range width.
User describes their maintenance context, including history such as postpartum changes, menopause, surgery, long-term fluctuation patterns, or expected long-term GLP-1 use.
User logs weigh-ins, dose events, and workout context to explain trend changes.
User reviews a dashboard that translates recent behavior into range status, 30-day fluctuation, and month-over-month change.
User uses dose tracking to see medication context alongside maintenance outcomes, without the app pretending to prescribe care.
User eventually syncs data and exports a summary when they want to share context with a coach or clinician.

Native onboarding for target weight and range setup
Default 5 lb maintenance band with customizable range width
Lightweight journey profile with narrative history plus a few structured maintenance-context fields
Local persistence for profile, weights, dose events, and workouts
Dashboard that shows:
- current in-range or out-of-range status
- 30-day fluctuation summary
- month-over-month fluctuation comparison
Logging, trends, activity, and profile views
GLP-1 dose logging framed as descriptive maintenance context, not as generic medication data entry or dosage optimization
Backend scaffold for auth, sync, imports, exports, and recommendations

HealthKit write flows and more automated import behavior
Stronger interpretation layers that connect dose, activity, and fluctuation over time
Coach or clinician-specific summary views

Tech: SwiftUI + SwiftData on iOS, Cloudflare Workers + Neon for backend services
Legal/compliance: avoid making unsupported clinical claims; preserve user-controlled sharing
Data/privacy: private-by-default posture, especially around health-adjacent data

Dashboards: activation funnel, weekly retention, logging cadence, fluctuation-view engagement, export usage

What is the minimum viable sync model for launch: account-required or optional account linking after local use?
Which fluctuation calculation is most trustworthy and explainable in a first release?
Should journey context start as one long freeform note, or a hybrid of note plus structured prompts?