OpenClaw Gateway: Configuration Reference

This guide covers OpenClaw gateway configuration settings and best practices for production deployments.

Gateway Configuration

The OpenClaw gateway configuration file controls how your AI assistant connects to channels, agents, and external services. Here are the key settings:

{
  "port": 18789,
  "mode": "local",
  "bind": "lan",
  "controlUi": {
    "allowedOrigins": [
      "http://localhost:18789",
      "http://127.0.0.1:18789"
    ]
  },
  "auth": {
    "mode": "token"
  }
}

Agent Defaults

Configure default model providers and behavior settings for your agents:

{
  "defaults": {
    "model": {
      "primary": "[PRIMARY_MODEL]",
      "fallbacks": ["[FALLBACK_MODEL_1]", "[FALLBACK_MODEL_2]"]
    }
  }
}

Channel Setup

OpenClaw supports multiple communication channels including Telegram, Discord, and Signal. Each channel requires its own API credentials configured in the gateway settings.

Configuration examples generalized for security. Specific values depend on your deployment.

Want to run OpenClaw on your own VPS? Vultr offers $100 in credit for new users — deploy an Ubuntu or Debian server in 60 seconds. Get started with Vultr →

Secure your OpenClaw VPS connection: NordVPN encrypts your traffic and protects your server from SSH attacks. Try NordVPN →

OpenClaw Gateway: Configuration Reference