🧠 Second Brain

🚀 What is Second Brain?

Second Brain v5.0 is a 100% local AI-powered personal knowledge management system with multimodal search, Google Drive integration, and photo processing capabilities. NO API KEYS REQUIRED for core functionality - runs entirely on local models (CLIP, LLaVA, Whisper).

📊 Latest Achievements

64,000+ Photos Processed - Google Takeout photo pipeline
100% Local AI - No cloud dependencies for core features
Multimodal Search - Text, image, and semantic search
Real-time Processing - WebSocket support for live updates
768-dim CLIP Embeddings - State-of-the-art image understanding

✨ Key Features

📸 Photo Pipeline

Process Google Takeout exports with automatic HEIC conversion, CLIP embeddings, and LLaVA descriptions.

🔍 Multimodal Search

Search by text, image similarity, or semantic meaning across all your memories.

🏠 100% Local Option

Run entirely on local models - CLIP for embeddings, LLaVA for vision, Whisper for audio.

🗄️ PostgreSQL + pgvector

Unified storage for vectors, metadata, and relationships with sub-100ms search.

📁 Google Drive Integration

OAuth 2.0 integration for automatic document and photo sync.

📊 Real-time Dashboard

WebSocket-powered live metrics, processing status, and performance monitoring.

💻 Quick Start

# Clone the repository
git clone https://github.com/raold/second-brain.git
cd second-brain

# Set up Python environment
python -m venv .venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate
pip install -r requirements.txt

# Copy environment template
cp .env.example .env

# Start full stack with Docker (recommended)
docker-compose up -d

# Or run development server
python scripts/run_dev.py  # Port 8001, uses SQLite/mock

# For photo processing
python process_all_photos.py  # Process Google Takeout archives

📚 API Usage

Create a Memory

curl -X POST http://localhost:8000/api/v2/memories \
  -H "Content-Type: application/json" \
  -d '{
    "content": "Important meeting notes from today",
    "importance_score": 0.8,
    "tags": ["work", "meeting"]
  }'

Search Memories

curl http://localhost:8000/api/v2/memories/search?query="meeting"

Get Statistics

curl http://localhost:8000/api/v2/statistics

🏗️ Architecture

Second Brain v5.0 features a comprehensive AI stack:

Core Services

PostgreSQL + pgvector - Vector database (port 5432)
FastAPI - Main API server (port 8000)
CLIP Service - Image embeddings (port 8002)
LLaVA (LM Studio) - Vision understanding (port 1234)

Optional Enhancements

OpenAI/Anthropic - Enhanced embeddings and generation
Whisper - Audio transcription
Google Drive - Cloud storage integration

📦 Project Structure

second-brain/
├── app/                    # Main application
│   ├── core/              # Infrastructure
│   ├── models/            # Data models
│   ├── routes/            # API endpoints
│   ├── services/          # Business logic
│   └── storage/           # Database backends
├── tests/                 # Test suites
├── docs/                  # Documentation
├── scripts/               # Utility scripts
├── docker/                # Docker configs
└── frontend/              # Web UI (SvelteKit)

🔗 Resources

🤝 Contributing

We welcome contributions! Please see our Contributing Guide for details.

📝 License

This project is licensed under the GNU Affero General Public License v3.0 (AGPL-3.0).

Made with ❤️ by the Second Brain Team