How rahulgpi/atharva_wetried works
Project type and purpose
• This is a full-stack web application combining a React-based frontend (“frontend/village”) with a Python/Flask backend (“backend/”).
• Purpose: provide a gamified, AI-driven mental wellness companion (virtual dog), where real-life tasks yield in-game rewards and AI-powered dialogue via the Gemini engine.
File structure
• Top level splits into two folders:
• backend/ – Flask service modules and requirements.txt
• frontend/village/ – React + Vite + Tailwind client code, public assets, config files
• Backend files indicate modular separation of concerns:
• User.py – user model or endpoints
• game.py – game logic
• survey.py – DASS-21 psychiatric survey logic
• avatar.py – companion state and assets
• gemini.py – AI service wrapper
• lifeline_manager.py – in-game lifeline or assistance functions
• config.py – configuration (DB URI, Gemini credentials)
• main.py and app.py – Flask app entrypoints
• Frontend has:
• src/pages/ – React page components (VillagePage, DogTTSAssistant, LangPage)
• src/assets/ & public/assets – audio, questions.json, static files
• package.json, vite.config.js, eslint.config.js – build & lint setup
README for additional insights
• Tech stack: React + Tailwind CSS, Flask, Gemini AI, MongoDB
• Features: gamification, AI therapy, real-life task ↔ in-game reward flow
• External dependencies: runs MongoDB locally or via cloud, calls Gemini AI API
a. Main components
• Client (React/Vite/Tailwind)
• Backend API (Flask service)
• AI Service (Gemini API)
• Database (MongoDB)
• Static Asset Store (public assets, audio, JSON) – could be a CDN or S3
b. Component relationships & data flow
1. User interacts in browser → React UI
2. React sends HTTP(S) requests (REST or GraphQL) to Flask API endpoints (/user, /game, /survey, /avatar, /lifeline)
3. Flask controller invokes business logic modules (User, Game, Survey, Avatar, LifelineManager)
4. Business logic may:
– Query/Update MongoDB via config’d client
– Call Gemini API via gemini.py to generate AI responses
5. Flask returns JSON responses to the frontend
6. React updates state, renders UI, plays audio/assets from public folder
c. Architectural patterns & design principles
• Modular monolith on backend (clear separation of domain modules)
• Single-page application (SPA) on frontend
• RESTful API layer between client and server
• External service integration (Gemini AI)
• Store static assets and configuration separately
d. Key technologies & frameworks
• React, Vite, Tailwind CSS – frontend
• Flask, Python – backend API
• Gemini AI – external LLM service
• MongoDB – document database
• Node.js (dev) – frontend build
• Virtual environment & pip – backend environment