🚀 Quick Start Guide¶

Get up and running with EasyPanel MCP in 5 minutes and deploy your first service using AI!

⏱️ 5-Minute Setup¶

Step 1: Install (1 minute)¶

# Clone repository
git clone https://github.com/dannymaaz/easypanel-mcp
cd easypanel-mcp

# Create virtual environment and activate
python -m venv venv
source venv/bin/activate  # Windows: venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

Step 2: Configure (1 minute)¶

# Copy environment file
cp .env.example .env

# Edit .env with your settings
# Use your favorite editor: nano, vim, notepad, etc.

Edit these two lines in .env:

EASYPANEL_URL=https://your-easypanel.com
EASYPANEL_API_KEY=your_api_key_here

Getting Your API Key

Open your EasyPanel dashboard
Go to Settings → API Keys
Click Generate New Key
Copy the key and paste it in .env

Step 3: Test Connection (1 minute)¶

# Quick connection test
python -c "
import asyncio
from src.client import EasyPanelClient
from config import config

async def test():
    client = EasyPanelClient(config.easypanel)
    await client.connect()
    healthy = await client.health_check()
    print(f'✅ EasyPanel connection: {\"OK\" if healthy else \"FAILED\"}')
    await client.disconnect()

asyncio.run(test())
"

Expected output:

✅ EasyPanel connection: OK

Step 4: Start Server (30 seconds)¶

# Start MCP server
python src/server.py

The server is now running and waiting for AI agent commands!

Step 5: Connect AI Agent (2 minutes)¶

Choose your AI agent:

Claude Desktop

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "easypanel": {
      "command": "python",
      "args": ["/path/to/easypanel-mcp/src/server.py"],
      "env": {
        "EASYPANEL_URL": "https://your-easypanel.com",
        "EASYPANEL_API_KEY": "your_key"
      }
    }
  }
}

Then restart Claude Desktop.

HTTP Mode (n8n, custom)

# Start in HTTP mode
python src/server.py http

Then send requests to http://localhost:8080/mcp

💬 Your First AI Deployment¶

Now you can talk to your AI agent naturally!

Example Conversation¶

Deploy a Flask API with PostgreSQL

You: "Deploy my Flask API with PostgreSQL database"

AI Agent:

I'll help you deploy a Flask API with PostgreSQL. Let me:

1. Create a new project called 'flask-api'
2. Deploy PostgreSQL database
3. Deploy Flask application
4. Configure networking

Starting deployment...

✅ Project 'flask-api' created
✅ Service 'postgres-db' deployed (postgres:15)
✅ Service 'flask-app' deployed (your-flask:latest)
✅ Internal network 'flask-net' created
✅ Services connected securely

🎉 Deployment complete!
📊 Dashboard: https://flask-api.easypanel.com
🔗 API: https://api.flask-api.easypanel.com

🎯 Common Commands¶

Here are some natural language commands you can use:

Service Management¶

"List all my services"
"Show me the logs for the worker service"
"Restart the API service"
"Create a new Redis cache service"

Project Management¶

"Create a new project for my staging environment"
"Show me all projects"
"Delete the test project"

Network Configuration¶

"Create an internal network for my database"
"List all networks"
"Set up isolated networking for my services"

Deployments¶

"Deploy version 2.0 of my application"
"Show me the deployment history"
"Rollback to the previous version"

📋 Complete Example: Full Stack App¶

Let's deploy a complete full-stack application:

Request¶

"Deploy my full-stack app with React frontend, Node.js backend, 
PostgreSQL database, and Redis cache. Use an internal network 
for the database and cache."

What the AI Will Do¶

1. ✅ Create project 'fullstack-app'
2. ✅ Create internal network 'app-internal'
3. ✅ Deploy PostgreSQL service (postgres:15)
   - Connected to app-internal network
   - Environment variables configured
4. ✅ Deploy Redis service (redis:7-alpine)
   - Connected to app-internal network
5. ✅ Deploy Node.js backend
   - Connected to app-internal and public networks
   - Database connection configured
6. ✅ Deploy React frontend
   - Public access enabled
   - Backend URL configured

🎉 Full stack deployed successfully!

🔍 Verify Your Deployment¶

Check Services¶

"Show me all running services"

AI Response:

{
  "success": true,
  "data": [
    {"name": "postgres-db", "status": "running", "image": "postgres:15"},
    {"name": "redis-cache", "status": "running", "image": "redis:7-alpine"},
    {"name": "nodejs-backend", "status": "running", "image": "myapp/backend:latest"},
    {"name": "react-frontend", "status": "running", "image": "myapp/frontend:latest"}
  ],
  "message": "Found 4 services"
}

Check Logs¶

"Show me the last 50 logs from the backend"

Health Check¶

"Is everything running healthy?"

🛠️ Next Steps¶

Now that you have the basics:

Learn More¶

Tools Reference - All available tools
Integration Guides - Connect your AI
Advanced Examples - Complex workflows

Best Practices¶

Security - Secure your setup
Networking - Isolate services
Monitoring - Monitor deployments

🆘 Troubleshooting¶

Server Won't Start¶

Connection refused to EasyPanel

Solution: 1. Verify EASYPANEL_URL is correct 2. Check API key is valid 3. Ensure EasyPanel is running

AI Agent Can't Connect¶

MCP server not responding

Solution: 1. Check server is running: python src/server.py 2. Verify path in AI agent config 3. Check for port conflicts

Deployment Fails¶

Service creation failed

Solution: 1. Check EasyPanel permissions 2. Verify project exists 3. Review service configuration

📚 Quick Reference¶

Environment Variables¶

EASYPANEL_URL=https://your-easypanel.com
EASYPANEL_API_KEY=your_key
MCP_PORT=8080

Start Commands¶

# Standard mode (stdio)
python src/server.py

# HTTP mode
python src/server.py http

Test Connection¶

python src/server.py  # Then send MCP requests

✅ Success Checklist¶

🎉 Congratulations! You're now deploying with AI.
Explore Tools Reference for all capabilities.