README.md

metadata

title: Second Brain
emoji: 🧠
colorFrom: indigo
colorTo: purple
sdk: gradio
sdk_version: 6.6.0
python_version: '3.10'
app_file: app.py
pinned: false

🧠 Second Brain

An intelligent personal productivity app that learns how you work and schedules your day accordingly.

Features

• Task Capture — type or speak tasks; AI classifies them across 4 dimensions automatically
• Life Areas — every task belongs to one of your personal life areas (Work, Health, etc.)
• Smart Scheduling — AI builds a time-blocked daily plan based on your tasks, context, and learned patterns
• Habit Tracking — mark tasks as habits with a recurrence interval; they auto-spawn each day
• Goals — define big-picture goals; AI factors them in when prioritizing
• Journaling — end-of-day reflection that updates the AI's knowledge of how you work best

Project Structure

second_brain/
├── app.py                  ← Main Gradio app, all UI + event wiring
├── requirements.txt
├── core/
│   ├── __init__.py
│   ├── database.py         ← SQLite: users, tasks, life areas, goals, AI context
│   ├── ai_engine.py        ← All Groq API calls (task parsing, scheduling, journaling)
│   └── styles.py           ← CSS
└── ui/
    └── __init__.py

---

Deploying to HuggingFace Spaces

Add your Groq API key as a Secret

In your Space on HuggingFace:

1. Go to Settings → Repository secrets
2. Click New secret
3. Name: GROQ_API_KEY
4. Value: your Groq key (starts with gsk_)
5. Click Add secret

⚠️ Never paste your API key directly into any file — use secrets only.

---

Notes

• Database: SQLite file second_brain.db is created automatically on first run in the app's working directory
• Voice transcription: Uses OpenAI Whisper small model (~250MB), downloaded on first voice task
• AI model: Groq llama-3.3-70b-versatile (free tier, ~0.3s response time)
• Persistence on HF Spaces: HuggingFace Spaces have ephemeral storage — the SQLite DB resets on each restart. For permanent storage, upgrade to a paid Space with persistent storage, or swap database.py to use an external DB (Supabase, PlanetScale, etc.)