Set Up a Discord Bot with OpenClaw: Step-by-Step Guide

Introduction

Discord is no longer just for gamers. More and more businesses, communities, and open-source projects use Discord as their primary communication platform. With OpenClaw, you can add an AI bot to your Discord server that answers questions, summarizes discussions, or serves as a knowledge base — all powered by the LLM model of your choice.

In this article, we walk through the complete process: from creating a Discord application and bot token to configuring the connection in OpenClaw and setting up permissions on your server.

Creating a Discord Application and Bot

Go to the Discord Developer Portal at discord.com/developers/applications and click "New Application". Give your application a name — this is the name that appears in audit logs, not the bot's display name. Then navigate to the "Bot" tab and click "Add Bot".

Under the "Bot" tab, you can set the bot's username and profile picture. Important: enable the "Message Content Intent" under "Privileged Gateway Intents". Without this setting, your bot cannot read message content and OpenClaw will be unable to respond to questions.

Copy the bot token via the "Reset Token" button. You will use this token in OpenClaw. Store it securely — if you lose the token, you need to reset it, which invalidates the old token.

Inviting the Bot to Your Server

Go to the "OAuth2" tab and select the scopes "bot" and "applications.commands" under "URL Generator". Under "Bot Permissions", select at minimum: "Send Messages", "Read Message History", and "Use Slash Commands". Copy the generated URL and open it in your browser to add the bot to your Discord server.

After inviting, the bot appears as an offline member on your server. This is normal — the bot only comes online once OpenClaw establishes the connection using the bot token.

Configuration in OpenClaw

Log in to your OpenClaw dashboard and go to "Channels". Choose "Add Discord" and enter the bot token you copied earlier. OpenClaw verifies the connection and displays your bot's name. Then select an LLM model and set a system prompt that fits the use case on your Discord server.

A useful setting for Discord is configuring a prefix or mention trigger. You can set the bot to only respond when directly mentioned with @BotName, or you can have it respond to every message in a specific channel. This prevents the bot from responding unwantedly in busy channels.

Conclusion

With OpenClaw, you add an intelligent AI assistant to your Discord server in under fifteen minutes. The combination of Discord as a communication platform and a powerful LLM model as a backend offers endless possibilities — from community support and FAQ answering to internal knowledge sharing and project documentation.

In a follow-up article, we will cover advanced Discord settings, such as using slash commands and restricting the bot to specific channels.