MindGarden

MindGarden is a lightweight habit and daily check-in web application that helps users build consistency by tracking habits, logging daily mood and reflections, computing streak-based insights, and generating short AI suggestions using recent history (optionally RAG-backed with embeddings and FAISS).

---

Features

• Email/password authentication (token-based)
• Habit management (create, list, delete)
• Free-tier rule: maximum 3 habits
• Daily check-ins:
  • Mood score
  • Optional reflection note
  • Per-habit completion tracking
  • Enforces one check-in per day
• Insights engine:
  • Habit streaks
  • 7-day mood average
• AI suggestions:
  • 1–2 sentence personalized “tiny challenge”
  • Optional deep-dive endpoint (premium hook)
• Reflection embeddings (RAG):
  • Stores embedded reflections
  • FAISS-based retrieval of relevant past notes
  • Graceful fallback if embeddings are unavailable
• Observability:
  • Metrics endpoint with daily counters and latency stats
• Monetization hooks:
  • subscription_tier on users
  • Upgrade endpoint (no payments yet)
  • Reflection export hook
• Demo seeding:
  • Endpoint creates a deterministic demo user, habits, and 7 days of data for recruiter demos

---

Tech Stack

• FastAPI
• React + Vite
• PostgreSQL
• SQLAlchemy
• Docker & Docker Compose
• Caddy reverse proxy
• FAISS + sentence-transformers (optional)
• pytest
• Python 3.11

---

Requirements

• Docker Desktop
• Git
• Optional: Python 3.11+

---

Getting Started (Windows / PowerShell)

Run the following commands in order:

git clone https://github.com/your-username/mindgarden.git
cd mindgarden

Copy-Item .env.example .env

docker compose up -d --build

---

Accessing the App

App: http://localhost:8080
API Docs (Swagger): http://localhost:8080/docs
OpenAPI JSON: http://localhost:8080/openapi.json

---

Demo Flow (Recruiter-Ready)

Seed demo data by calling:

Invoke-RestMethod -Method POST http://localhost:8080/api/dev/seed_demo

Then:

1. Log in with the demo user
2. Create today’s check-in live
3. View insights and AI suggestions
4. Show metrics and analytics endpoints

---

Running Tests

docker compose exec api pytest -vv

---

Project Structure

app/
main.py
models.py
db.py
auth/
habits/
checkins/
insights/
ai/
rag/
metrics/

frontend/
src/
index.html

worker/

tests/

docker-compose.yml
Dockerfile
Caddyfile
.env.example
README.md

---

Next Steps (Optional Enhancements)

• Real payment integration
• Richer analytics visualizations
• PDF / Markdown exports
• Mobile-friendly UI
• Background retraining of embeddings