How to Install OpenClaw (Moltbot/Clawdbot) on Hostinger VPS

OpenClaw (formerly known as Moltbot/Clawdbot) is a personal AI assistant platform that runs on your own infrastructure, enabling you to interact with AI across multiple messaging channels, including WhatsApp, Telegram, Slack, Discord, and more. This guide will walk you through deploying OpenClaw on your Hostinger VPS using the one-click Docker template.

Installation Steps

Deploy OpenClaw (If You Don’t Have a VPS Yet)

If you don’t have a VPS already, you can deploy OpenClaw directly from the marketplace:

1. Head to the OpenClaw VPS hosting page
2. Select your hosting plan that fits your needs
3. Click Deploy – the OpenClaw application will be automatically pre-selected
4. Complete the purchase

After completing the purchase, you’ll see a configuration screen where you can set up your environment variables:

Configuration Options:

• OPENCLAW_GATEWAY_TOKEN: This will be auto-generated for you
• ANTHROPIC_API_KEY: Your Anthropic API key for Claude integration (optional)
• OPENAI_API_KEY: Your OpenAI API key for AI integration (optional)

Once the installation is complete, a Docker project will be created automatically. This typically takes several minutes. After it’s done, you’ll see the OpenClaw application with exposed ports ready for access.

Install Docker Manager (If You Already Have a VPS)

If you already have an existing VPS, follow these steps:

Step 1: Install Docker Manager

1. Once your VPS is active, access the hPanel
2. From the left sidebar, locate and click on Docker Manager
3. If Docker Manager is not yet installed, click the installation button to set it up
4. Wait for the installation to complete (this typically takes 2-3 minutes)

Step 2: Deploy OpenClaw from the Catalog

1. In Docker Manager, navigate to the Catalog section
2. Use the search bar or browse the categories to find OpenClaw
3. Click on the OpenClaw card to view the application details
4. Click the Deploy button

Step 3: Configure Environment Variables

During deployment, you’ll be prompted to configure several environment variables:

Required:

• OPENCLAW_GATEWAY_TOKEN: This is auto-generated for you. This token will be used to access your OpenClaw web interface. Save this token securely.

Optional (for AI integration):

• ANTHROPIC_API_KEY: Claude web session key for browser-based authentication
• OPENAI_API_KEY: Claude web cookie for session persistence

You can leave the optional fields empty and configure them later if needed.

Step 4: Complete the Deployment

1. Review your configuration
2. Click Deploy to start the installation
3. Docker will pull the OpenClaw image and start the container
4. Wait for the deployment status to show as “Running” (usually takes 1-2 minutes)

Accessing OpenClaw

Once deployed, you can access your OpenClaw instance:

1. In Docker Manager, navigate to your deployed OpenClaw project
2. Note the assigned port number for the web interface
3. Access OpenClaw at: http://your-vps-ip:port or by simply pressing on the port in the project.

   

Post-Installation Configuration

Step 1: Enter Your Gateway Token

1. In the OpenClaw web interface, go to Overview
2. You’ll see a field for the Gateway Token
3. Insert the OPENCLAW_GATEWAY_TOKEN that you copied during the installation step
4. Press on Connect to authenticate your access, you will see “Connected” on the right.

Step 2: Connecting Messaging Channels

After your OpenClaw instance is running, you can connect it to various messaging platforms:

1. Navigate to the Settings -> Config section
2. Click on the RAW button to access the raw configuration editor
3. Add the following configuration to enable WhatsApp:

"channels": {
  "whatsapp": {
    "dmPolicy": "allowlist",
    "allowFrom": [
      "[INSERT-YOUR-PHONE-NUMBER-HERE]"
    ],
    "groupPolicy": "allowlist",
    "mediaMaxMb": 50,
    "debounceMs": 0
  }
},

1. Click Apply to apply your configuration changes
2. Click Update to update your configuration
3. Go to Channels and press on Show QR

4. Head to your WhatsApp Settings -> Linked devices

Press on “Wait for scan” and scan the code with your WhatsApp app by pressing on “Link a device“.

Step 3: Configure AI Model (Optional)

If you’re using the OpenAI API, you’ll need to configure your preferred model, because OpenClaw doesn’t use it as the default:

1. In the same RAW configuration editor
2. Add the following configuration to set up your OpenAI agent:

"agents": {
  "defaults": {
    "model": {
     "primary": "openai/gpt-4o"
    },
  "maxConcurrent": 4,
  "subagents": {
  "maxConcurrent": 8
  }
 }
},