API Reference
Start typing to search…

Discord Bot using APIs

A Discord bot that integrates with CustomGPT's API to answer questions using your agent's knowledge base.

More information on GitHub - https://github.com/Poll-The-People/customgpt-integrations

Features

  • 🤖 AI-Powered Responses: Uses CustomGPT's API to provide intelligent answers
  • 💬 Conversation Memory: Maintains context within Discord channels
  • 🚀 Starter Questions: Interactive buttons for common questions
  • 📚 Source Citations: Shows sources for answers when available
  • ⏱️ Rate Limiting: Prevents abuse with configurable limits
  • 🔒 Security: Channel and role-based access control
  • 📄 Pagination: Handles long responses elegantly
  • ⚡ Real-time: Typing indicators and responsive UI

Setup Instructions

1. Prerequisites

2. Installation

# Clone the repository
git clone <your-repo-url>
cd discord

# Install dependencies
pip install -r requirements.txt

# Copy environment variables
cp .env.example .env

3. Configuration

Edit .env file with your credentials:

# Required
DISCORD_BOT_TOKEN=your_discord_bot_token
CUSTOMGPT_API_KEY=your_customgpt_api_key
CUSTOMGPT_AGENT_ID=your_agent_id

# Optional (defaults shown)
DISCORD_COMMAND_PREFIX=!
RATE_LIMIT_PER_USER=10
RATE_LIMIT_PER_CHANNEL=30

4. Discord Bot Setup

  1. Go to Discord Developer Portal
  2. Create a New Application
  3. Go to Bot section and create a bot
  4. Copy the bot token
  5. IMPORTANT: Enable Privileged Gateway Intents:
    • Scroll down to "Privileged Gateway Intents"
    • Toggle ON the "MESSAGE CONTENT INTENT" (Required for the bot to read messages)
    • Click "Save Changes"
  6. Go to OAuth2 > URL Generator
  7. Select scopes: bot, applications.commands
  8. Select permissions: Send Messages, Embed Links, Read Message History, Use Slash Commands
  9. Use the generated URL to invite the bot to your server

5) Running Locally

python bot.py

Deployment Options

Option 1: Railway (Recommended - Free Tier Available)

  1. Install Railway CLI:
npm install -g @railway/cli
  1. Deploy:
railway login
railway init
railway up
  1. Set environment variables in Railway dashboard

Option 2: Fly.io (Free Tier Available)

  1. Install Fly CLI:
curl -L https://fly.io/install.sh | sh
  1. Deploy:
flyctl launch
flyctl deploy
flyctl secrets set DISCORD_BOT_TOKEN=xxx CUSTOMGPT_API_KEY=xxx CUSTOMGPT_AGENT_ID=xxx

Option 3: Replit (Free with Limitations)

  1. Create new Python repl on Replit
  2. Upload all files
  3. Set environment variables in Secrets tab
  4. Run the bot
  5. Use UptimeRobot to keep it alive

Option 4: Heroku (No longer free)

See heroku/ directory for Heroku-specific files.

Option 5: VPS (Self-hosted)

  1. Use systemd service (see discord.service file)
  2. Or use Docker (see Dockerfile)

Bot Commands

  • !ask [question] - Ask a question to the CustomGPT agent
  • !help - Show interactive help menu
  • !info - Display agent information
  • !starters - Show starter questions with buttons
  • !reset - Reset conversation in current channel

Advanced Configuration

Rate Limiting

Configure in .env:

  • RATE_LIMIT_PER_USER: Queries per minute per user
  • RATE_LIMIT_PER_CHANNEL: Queries per minute per channel
  • REDIS_URL: Optional Redis URL for distributed rate limiting

Access Control

Configure in .env:

  • ALLOWED_CHANNELS: Comma-separated channel IDs
  • ALLOWED_ROLES: Comma-separated role IDs

Redis Setup (Optional)

For distributed rate limiting across multiple bot instances:

  1. Free Redis: Redis Cloud (30MB free)
  2. Set REDIS_URL in .env

Monitoring

  • Check logs for errors
  • Monitor rate limit hits
  • Track API usage in CustomGPT dashboard

Troubleshooting

Bot not responding

  • Check bot token is correct
  • Verify bot has proper permissions
  • Check allowed channels/roles configuration

PrivilegedIntentsRequired Error

  • Go to Discord Developer Portal
  • Select your application > Bot section
  • Enable "MESSAGE CONTENT INTENT" under Privileged Gateway Intents
  • Save changes and restart the bot

API errors

  • Verify CustomGPT API key and Agent ID
  • Check API rate limits
  • Ensure agent is active

Rate limiting issues

  • Adjust limits in .env
  • Consider using Redis for better rate limiting

Security Best Practices

  1. Never commit .env file
  2. Use environment variables in production
  3. Regularly rotate API keys
  4. Monitor for unusual activity
  5. Set appropriate rate limits
  6. Use channel/role restrictions

Support

For issues related to:

  • Discord Bot: Check Discord.py documentation
  • CustomGPT API: Contact CustomGPT support
  • This integration: Open an issue on GitHub