とりメモ (Torimemo) - Blazingly Fast Bookmark Manager

A lightning-fast, lightweight bookmark manager with AI-powered auto-categorization, built with Go + TypeScript + SQLite.

✨ Features

• 🚀 Blazingly Fast: Single Go binary with embedded frontend
• 🪶 Ultra Lightweight: <50MB RAM usage, minimal resource footprint
• 🔍 Full-Text Search: Powered by SQLite FTS5 with ranked results
• 🏷️ Smart Tagging: Interactive tag cloud with auto-categorization
• 🎨 Cyberpunk UI: Futuristic design with neon effects
• ⚡ Real-Time: Instant search and updates
• 🐳 Container Ready: Docker & Kubernetes deployment configs
• 🤖 AI Ready: Learning system for intelligent categorization

🚀 Quick Start

Local Development

# Clone and build
git clone <repo>
cd bookmark.arcbjorn.com
./build.sh

# Run
./torimemo

# Open http://localhost:8080

Docker

# Build and run
docker build -t torimemo .
docker run -p 8080:8080 -v torimemo_data:/data torimemo

# Or use docker-compose
docker-compose up -d

Kubernetes

kubectl apply -f k8s.yaml

📖 API Documentation

Bookmarks

• GET /api/bookmarks - List bookmarks (with pagination, filtering)
• POST /api/bookmarks - Create bookmark
• GET /api/bookmarks/{id} - Get bookmark
• PUT /api/bookmarks/{id} - Update bookmark
• DELETE /api/bookmarks/{id} - Delete bookmark
• GET /api/bookmarks/search?q={query} - Full-text search

Tags

• GET /api/tags - List tags
• GET /api/tags/cloud - Tag cloud with sizes
• GET /api/tags/popular - Popular tags
• POST /api/tags - Create tag
• PUT /api/tags/{id} - Update tag
• DELETE /api/tags/{id} - Delete tag

System

• GET /api/health - Health check
• GET /api/stats - Database statistics

🏗️ Architecture

• Backend: Go 1.25 (standard library only)
• Database: SQLite with FTS5, WAL mode
• Frontend: TypeScript + Lit Web Components + TailwindCSS
• Search: Full-text search with ranking and snippets
• Tags: Many-to-many with weighted tag cloud
• Deployment: Single binary with embedded assets

🎯 Performance

• Binary Size: 12MB (with embedded frontend + AI)
• Memory Usage: <50MB RAM
• Cold Start: <100ms
• Search Latency: <10ms for 10k+ bookmarks
• Database: SQLite with FTS5 indexing

🛠️ Development

# Frontend development
cd web
pnpm install
pnpm run dev

# Backend development  
go run main.go

# Build optimized
./build.sh

📦 Deployment Options

Single Binary Deployment

The simplest deployment - just copy the binary:

# Build single binary (includes frontend)
./build.sh

# Deploy anywhere - single 12MB binary
scp torimemo user@server:/usr/local/bin/
ssh user@server 'torimemo'

Production Deployment

Quick production setup with SSL and reverse proxy:

# Copy and edit environment variables
cp .env.example .env.production
# Edit .env.production with your values

# Deploy with SSL-enabled reverse proxy
./deploy.sh

Environment Variables

• PORT: Server port (default: 8080)
• DB_PATH: Database path (default: ./torimemo.db)
• LOG_LEVEL: Logging level (DEBUG, INFO, WARN, ERROR)
• ALLOWED_ORIGINS: CORS allowed origins
• RATE_LIMIT: Rate limit (requests per minute)
• RATE_BURST: Rate limit burst size
• JWT_SECRET: JWT signing secret (change in production!)
• JWT_EXPIRY: JWT token expiry duration

Docker Compose

# Development
docker-compose up -d

# Production with nginx reverse proxy
docker-compose -f docker-compose.prod.yml up -d

Production Checklist

• Set strong JWT_SECRET in production
• Configure ALLOWED_ORIGINS for your domain
• Use real SSL certificates (replace self-signed)
• Set up database backups
• Configure log rotation
• Monitor health endpoint at /api/health
• Set appropriate LOG_LEVEL (WARN for production)

🤖 AI Categorization

Three-layer AI system for intelligent bookmark categorization:

1. Rule-based: Domain patterns, URL analysis (always active)
2. FastText: Lightweight text classification (optional)
3. ONNX Models: Advanced content understanding (optional)

Quick AI Setup

# Create model directories
mkdir -p ./models/fasttext ./models/onnx

# Start application (Layer 1 works immediately)
./torimemo

# Check AI predictive analysis
curl -X POST http://localhost:8080/api/ai/predict-tags \
  -H "Content-Type: application/json" \
  -d '{"url":"https://github.com","title":"GitHub","description":"Code repository"}'

# Find duplicate bookmarks  
curl -X POST http://localhost:8080/api/ai/duplicates/check \
  -H "Content-Type: application/json" \
  -d '{"url":"https://github.com/golang/go","title":"Go Programming"}'

AI API Endpoints

• POST /api/ai/predict-tags - Get predictive tag suggestions
• POST /api/ai/duplicates/check - Check for duplicate bookmarks
• POST /api/ai/duplicates/find-all - Find all duplicate groups
• POST /api/ai/cluster - Cluster bookmarks by similarity
• POST /api/ai/cluster/analyze - Analyze clustering potential
• POST /api/ai/predict/learn - Submit learning feedback

📖 Full AI Setup Guide: See AI_SETUP.md for detailed model installation and configuration.

📄 License

MIT License - see LICENSE file for details.

---

Built with ❤️ in Go and TypeScript