Build Your Own Discord Bot

Build an AI-powered Discord bot using Devbox. You'll create a bot that can respond to messages using AI capabilities powered by FastGPT.

Overview

This example demonstrates how to build an AI-powered Discord bot using Devbox. You'll create a bot that can respond to messages using AI capabilities powered by FastGPT, showcasing how to integrate multiple services and APIs in a Devbox environment.

Technologies Used:

Node.js
Discord.js
FastGPT API
Devbox development environment

Expected Outcome:

A functioning Discord bot that:
- Responds to basic commands
- Integrates with FastGPT for AI-powered responses
- Handles real-time message events
- Processes natural language queries

Prerequisites

A Discord account with administrator privileges
Basic knowledge of JavaScript/Node.js
Access to FastGPT platform
The following credentials:
- Discord Bot Token
- FastGPT API Key
- FastGPT Base URL

Step-by-Step Guide

1. Setting Up Discord Application

Create Discord Application

Visit the Discord Developer Portal
Click "New Application" and choose a name for your bot. This name will be displayed in your Discord server

Configure Bot Permissions

Navigate to the "Bot" section
Enable required intents:
- Server Members Intent
- Message Content Intent These permissions allow the bot to:
  - Access member-related events
  - Read and process message content
Copy your bot token by clicking "Reset Token" and then "Copy"
- Keep this token secure and never share it
- You'll need this token to authenticate your bot

Set Up OAuth2

Go to OAuth2 section
Select "bot" scope
Choose "Administrator" permissions
Copy the generated OAuth2 URL
Use the URL to add the bot to your server

2. Creating Devbox Project

Initialize Node.js Project

Create a new Devbox project
Select Node.js as the runtime
Configure project resources:
- Set appropriate CPU cores
- Allocate required memory

Set Up Development Environment

Connect to Your Development Environment
Install required dependencies:

npm init -y
npm install discord.js axios dotenv

Update package.json to enable ES modules:

{
  "type": "module",
  "scripts": {
    "start": "node src/index.js"
  }
}

Project Structure

Create the following file structure:

project/
├── src/
│   ├── index.js
│   ├── services/
│   │   └── aiService.js
├── package.json
└── README.md

3. Implementing the Bot

Basic Bot Setup

Create src/index.js:

src/index.js

import { Client, GatewayIntentBits } from 'discord.js';
import 'dotenv/config';
import aiService from './services/aiService.js';
 
const client = new Client({
  intents: [
    GatewayIntentBits.Guilds,
    GatewayIntentBits.GuildMessages,
    GatewayIntentBits.MessageContent
  ]
});
 
client.on('ready', () => {
  console.log(`${client.user.tag} is ready!`);
});
 
client.on('messageCreate', async message => {
  if (message.author.bot) return;
  
  if (message.content === '!ping') {
    message.reply('Pong!');
  }
  
  if (message.content.startsWith('!ask ')) {
    const question = message.content.slice(5).trim();
    if (!question) {
      message.reply('Please enter your question after !ask. For example: !ask what is Sealos?');
      return;
    }
    
    try {
      message.channel.sendTyping();
      const response = await aiService.getChatGPTResponse(question);
      message.reply(response);
    } catch (error) {
      console.error('Error getting AI response:', error);
      message.reply('Sorry, I cannot answer right now. Please try again later.');
    }
  }
});
 
client.login(process.env.DISCORD_TOKEN);

AI Service Integration

Create src/services/aiService.js:

src/services/aiService.js

import axios from 'axios';
 
class AiService {
  constructor() {
    this.openaiAxios = axios.create({
      baseURL: process.env.FASTGPT_BASE_URL,
      headers: {
        'Authorization': `Bearer ${process.env.FASTGPT_API_KEY}`,
        'Content-Type': 'application/json'
      }
    });
  }
 
  async getChatGPTResponse(message) {
    try {
      const response = await this.openaiAxios.post('/v1/chat/completions', {
        chatId: "session_" + Date.now(),
        stream: false,
        detail: false,
        messages: [{
          role: "user",
          content: message
        }]
      });
      return response.data.choices[0].message.content.trim();
    } catch (error) {
      console.error('AI API call failed:', error);
      return 'Sorry, I cannot answer right now. Please try again later.';
    }
  }
}
 
export default new AiService();

Configuration Setup

Create .env file in the project root with your credentials:

DISCORD_TOKEN=your_discord_bot_token
FASTGPT_API_KEY=your_fastgpt_api_key
FASTGPT_BASE_URL=your_fastgpt_base_url

Update src/index.js to include dotenv configuration:

import 'dotenv/config';
import { Client, GatewayIntentBits } from 'discord.js';
// ... rest of the existing code ...

Start the bot with:

npm start

4. FastGPT Integration

FastGPT is an open-source LLM application development platform that allows you to build AI applications with knowledge base capabilities. In this section, we'll walk through the process of setting up FastGPT to power our Discord bot with custom knowledge.

Dataset Creation

Log into your FastGPT account at FastGPT
Navigate to the "Datasets" section and click "Create Dataset"
Choose "General Dataset" as the dataset type
Select "Text Dataset" for document processing
Upload your documentation files (supports PDF, TXT, Word, etc.)
Configure chunking settings if needed (default settings work well for most cases)
Wait for the training process to complete - this may take several minutes depending on the size of your documentation

Application Setup

Go to the "Applications" section and create a new FastGPT application
In the Flow Editor, add a "Dataset Search" module
Connect your trained dataset to the search module
Add an "AI Chat" module and connect it to the search module
Configure the AI Chat module settings:
- Set temperature (0.7 recommended for balanced responses)
- Adjust max tokens as needed
- Customize the system prompt

FastGPT Application Setup

API Publication

Once satisfied with the configuration, click "Publish Channel" to make your application live
Go to the "API Request" section of your application
Generate new API credentials if you haven't already
Save these important details for the next steps:
- API Base URL: The endpoint for your FastGPT API
- API Key: Your authentication token

These credentials will be used in the .env file we created earlier to connect our Discord bot to the FastGPT backend.

Build Your Own Discord Bot

On this page