DayMaxing

Habit apps die at the guilt stage — break a streak, delete the app. And any self-reported progress system collapses the moment it can be gamed.

People — starting with me — who respond to progression systems but see through fake ones.

1. Three stats, not one score.

   Intelligence, Strength, and Discipline mirror how people actually think about self-improvement; a single number flattens motivation.

2. Level-ups are weekly and earned.

   85% weekly completion to level up — forgiving enough to survive a bad day, strict enough to mean something.

3. Anti-cheat is designed into the database layer.

   XP is locked and auto-computed from cadence × completion type × target × penalty — the client can't claim progress, only report completions. The Postgres migrations (RLS, locked columns) are written; wiring them up is the current milestone.

Next.js 16 + React 19, TypeScript strict, Tailwind v4; motion for scroll animation, and three.js/react-three-fiber for a Mixamo-rigged character that shifts pose as you scroll — idle → thinking (INT) → push-ups (STR) → meditation (DIS). The XP, time, and quest logic is pure, unit-tested TypeScript (Vitest + Playwright e2e).

Persistence is localStorage today; the Supabase/Postgres migrations — schema, RLS, weekly planning — are written, and wiring them up is the current milestone.

The core loop is built and playable solo — weekly planning, three stats, locked XP, streaks with penalty quests. Hardening for multi-user is in progress, not finished.

Wire Supabase auth + sync — the migrations are ready. Then a daily morning quest email. Then cohorts.