Armony - Love Coach SaaS

Refonte de l application avec:

• Frontend moderne en React + Vite
• Backend Flask API stateless
• Integration Pollinations cote serveur (cle non exposee au client)
• Aucune persistance des conversations (alignement RGPD)

Product Status

Snapshot: March 4, 2026

Done:

• F1 Personalized 30-Day Plan
• F2 Daily Check-ins
• F3 Message Lab
• F4 Before-Send Mode
• UI SaaS: homepage, chat page, coach hub, message lab, legal pages, contact
• FR/EN switch and language-aware assistant behavior
• Pollinations backend integration (/api/chat)

In progress:

• F7 Playbooks v1

Remaining:

• F5 Conversation Simulator
• F6 Pattern Detection
• F8 Relationship Timeline
• F9 Couple Mode
• F11 Exercise Library
• Cross-sprint: analytics instrumentation, observability, rate limiting, E2E tests

Reference docs:

• docs/PRD.md
• docs/SPRINT_BACKLOG.md

Architecture

• src/: React interface
• app.py: Flask API (/api/chat, /api/health)
• .env.example: environment variables (backend + frontend)

Prerequisites

• Node.js 20+
• Python 3.11+

Installation

1. Install frontend dependencies:

npm install

2. Install backend dependencies:

pip install -r requirements.txt

3. Create .env from .env.example and set:

• POLLINATIONS_API_KEY
• POLLINATIONS_MODEL (optional, default openai)

Local Run

1. Start Flask backend:

python app.py

2. In another terminal, start frontend:

npm run dev

Frontend calls backend through VITE_API_BASE_URL.

Privacy / GDPR

• User chat messages are not stored in database.
• Backend does not keep persistent conversation history.
• App logs avoid message content.

Deployment

Procfile runs Flask API:

web: gunicorn app:app