​
OpenClaw Setup Guide

ClawBox ships with OpenClaw pre-installed, so most of this is already done for you. This guide is the full reference for configuring channels, voice, models, and advanced options — whether you’re on a ClawBox or a DIY install.

​
Prerequisites

  • Node.js 18+ installed (pre-installed on ClawBox)
  • 4 GB+ RAM recommended
  • Linux, macOS, or Windows
  • Always-on internet connection

​
Telegram

1

Create a bot

Open Telegram and search @BotFather.
2

Name it

Send /newbot and follow the prompts to create your bot.
3

Copy the token

BotFather gives you an API token — copy it.
4

Add it to OpenClaw

Paste the token into your OpenClaw config (Channels → Telegram).

​
Discord

1

Create an application

Go to the Discord Developer Portal and create a new application.
2

Create a bot

Add a bot to the application and copy its token.
3

Enable intents

Turn on the required gateway intents for the bot.
4

Add it to OpenClaw

Paste the token into your OpenClaw config (Channels → Discord).

​
Voice pipeline

  • Install Whisper for speech-to-text
  • Install Kokoro TTS for text-to-speech
  • Configure your audio devices

​
AI models

  • Add your API key (Anthropic, OpenAI, Google, OpenRouter), or download local models
  • Configure model preferences
  • Test with a simple prompt
See Choose Your AI Provider for the full provider list.

​
Advanced configuration

  • Set up systemd for auto-start (pre-configured on ClawBox)
  • Configure browser automation
  • Set up cron jobs for monitoring and scheduled tasks

​
Pro tips

  • Run openclaw status to check that all services are running.
  • Use systemctl restart openclaw if something misbehaves after a config change.
  • Tail logs in real time with journalctl -u openclaw -f.
  • Enable automatic updates for the latest features and security patches.

​
Skip the setup

Get ClawBox — OpenClaw pre-installed

Plug in, connect Wi-Fi, scan a QR code. Done in 5 minutes — channels, voice, and models ready.