🧠 OpenMemory MCP - Coolify Edition

Local-first AI Memory Layer with MCP Support

Your memories, your control. Privacy-focused AI memory that works across any MCP-compatible client.

🚀 Quick Start • 📖 Documentation • 🔌 MCP Setup • 🛠️ Development

---

✨ Features

• 🔒 Privacy First - All data stays on your server using official Mem0 setup
• 🌐 Universal Memory - Works with Claude, Cursor, Windsurf, and all MCP clients
• ⚡ Lightning Fast - Vector-based semantic search with Qdrant + PostgreSQL
• 🐳 One-Click Deploy - Uses Mem0's official easy setup script
• 🔄 Auto-Deploy - GitHub Actions integration for seamless updates
• 📊 Built-in Dashboard - Official Mem0 web UI for memory management
• 🏗️ Production Ready - Mem0's battle-tested architecture
• 📈 Monitoring Ready - Built-in health checks and logging

🚀 Quick Start

Option 1: One-Click Coolify Deployment

# Clone and run the installer
git clone https://github.com/yourusername/openmemory-mcp-coolify.git
cd openmemory-mcp-coolify
chmod +x coolify-install.sh

# Set required environment variables
export OPENAI_API_KEY="your-openai-api-key"
export DOMAIN="memory.yourdomain.com"

# Run the installer
./coolify-install.sh

Option 2: Manual Installation

# Clone repository
git clone https://github.com/yourusername/openmemory-mcp-coolify.git
cd openmemory-mcp-coolify

# Create environment file
cp .env.example .env
# Edit .env with your configuration

# Deploy with Docker Compose
docker-compose -f .coolify/docker-compose.yml up -d

📋 Prerequisites

• Docker & Docker Compose
• OpenAI API Key (Get one here)
• Domain name (optional, can use localhost)
• Coolify instance (for production deployment)

🛠️ Configuration

Environment Variables

Variable	Required	Default	Description
OPENAI_API_KEY	✅	-	OpenAI API key for embeddings
DOMAIN	❌	localhost	Your domain name
APP_NAME	❌	openmemory	Application name prefix
LOG_LEVEL	❌	INFO	Logging level (DEBUG, INFO, WARNING, ERROR)
MAX_MEMORY_SIZE	❌	1000	Maximum number of memories
EMBEDDING_MODEL	❌	text-embedding-3-small	OpenAI embedding model

🔌 MCP Client Setup

Claude Desktop

Add to your Claude Desktop configuration:

{
  "mcpServers": {
    "openmemory": {
      "transport": "sse",
      "url": "https://your-domain.com/mcp/claude/sse/YOUR_USERNAME"
    }
  }
}

Cursor

Add to your Cursor configuration:

{
  "mcpServers": {
    "openmemory": {
      "transport": "sse",
      "url": "https://your-domain.com/mcp/cursor/sse/YOUR_USERNAME"
    }
  }
}

Auto-Install (Recommended)

npx install-mcp i "https://your-domain.com/mcp/claude/sse/YOUR_USERNAME" --client claude

🚀 Coolify Deployment

1. Create New Project

1. Go to your Coolify dashboard
2. Create new project: "OpenMemory MCP"
3. Choose "Git Repository" as source
4. Set repository URL: https://github.com/yourusername/openmemory-mcp-coolify.git

2. Configuration

• Docker Compose File: .coolify/docker-compose.yml
• Environment: Set required variables
• Domain: Configure your domain with SSL

3. Deploy

Push to main branch or trigger manual deployment from Coolify dashboard.

📊 Services

Service	Port	Description	Health Check
UI	3000	React dashboard for memory management	GET /
API	8765	FastAPI backend with MCP endpoints	GET /health
Qdrant	6333	Vector database for semantic search	GET /health
Redis	6379	Optional caching layer	PING

🔧 Management Commands

# Deploy/Update
./scripts/deploy.sh

# Check status
./scripts/status.sh

# View logs
./scripts/logs.sh [service-name]

# Backup data
./scripts/backup.sh

# Stop services
./scripts/stop.sh

📝 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgments

• Mem0 - Original OpenMemory implementation
• Model Context Protocol - MCP specification
• Coolify - Self-hosted deployment platform

---

Made with ❤️ for the AI community

⭐ Star this repository if you find it helpful!