Skip to main content

​
Overview

Discord Webhook Manager includes AI-powered content generation to help you create engaging Discord messages quickly. The system supports two AI providers:
  • OpenAI GPT-4 (gpt-4o model)
  • Google Gemini (gemini-2.5-flash model)
AI generation is available in all message editors: Quick Send, Webhooks, and Templates.

​
Configuration

​
Enable AI Provider

Administrators can configure AI providers through the Settings panel:
1

Access Admin Settings

Navigate to the admin dashboard and select Settings
2

Choose AI Provider

Select either OpenAI or Gemini as your preferred provider
3

Add API Key

Enter your API key for the selected provider:
OPENAI_API_KEY=sk-...

​
Provider Configuration

Both providers are configured through the AiService class (app/app/Services/AiService.php:18): OpenAI Configuration:
  • Model: gpt-4o
  • Temperature: 0.7
  • Max tokens: 1000
  • Endpoint: https://api.openai.com/v1/chat/completions
Gemini Configuration:
  • Model: gemini-2.5-flash
  • Temperature: 0.7
  • Max output tokens: 1024
  • Endpoint: https://generativelanguage.googleapis.com/v1/models/gemini-2.5-flash:generateContent

​
User Access Control

​
Granting AI Access

Administrators control which users can access AI features through the user management panel:
  1. Navigate to Admin Dashboard → Users
  2. Toggle the AI Access switch for each user
  3. Set daily usage limits (default: 5 requests per day)
Administrators have unlimited AI usage and are exempt from daily limits.

​
Usage Limits

The system tracks AI usage per user through the ai_usages table (app/app/Controllers/AiController.php:26-37):
  • Non-admin users have a configurable daily limit (default: 5)
  • Usage resets daily at midnight
  • Administrators bypass all limits
  • Failed generations don’t count toward the limit

​
Using AI Generation

​
In Message Editors

The AI generation interface appears in all message composition screens:
1

Write Your Prompt

Enter a description of the content you want to generate:
Write a welcome message for new Discord members joining our gaming community
2

Generate Content

Click the Generate with AI button (sparkle icon)
3

Review and Edit

The generated content appears in the message editorYou can edit, refine, or regenerate as needed

​
Prompt Best Practices

Be specific in your prompts for better results:
  • Good: “Write an announcement about server maintenance on March 15th from 2-4 PM EST”
  • Poor: “Write something about maintenance”
The AI uses a specialized system prompt optimized for Discord content: 
Eres un experto en comunicación para Discord. Genera únicamente el 
contenido del mensaje solicitado utilizando formato Markdown de Discord 
(negritas, listas, emojis, etc.). No incluyas introducciones, 
explicaciones, ni textos adicionales.

​
Discord Markdown Support

The AI automatically formats content using Discord markdown:
  • Bold text: **text**
  • Italic text: *text*
  • Code blocks: `code`
  • Lists: - item
  • Emojis: Native Unicode emojis

​
API Response Handling

​
Error Scenarios

The AI controller handles various error cases (app/app/Controllers/AiController.php:14-62): Permission Denied (403) 
{
  "success": false,
  "message": "No tienes permiso para usar esta función."
}
Daily Limit Reached (429) 
{
  "success": false,
  "message": "Has alcanzado el límite diario de uso de la IA (5). ¡Vuelve mañana!"
}
Generation Failed (422) 
{
  "success": false,
  "message": "No se pudo generar el contenido. Esto puede deberse a una clave de API inválida, límites excedidos o un modelo no disponible."
}

​
Success Response

{
  "success": true,
  "content": "🎮 Welcome to our gaming community!\n\nWe're excited to have you here..."
}

​
Usage Tracking

Every successful AI generation is logged in the ai_usages table for analytics and quota management. Database Schema: 
CREATE TABLE ai_usages (
  id BIGINT PRIMARY KEY,
  user_id BIGINT NOT NULL,
  created_at TIMESTAMP,
  updated_at TIMESTAMP
);
Administrators can view usage statistics per user in the admin dashboard to:
  • Monitor AI adoption
  • Identify power users
  • Adjust daily limits as needed

​
Troubleshooting

​
API Key Issues

If generation fails, check the logs (storage/logs/laravel.log): 
OpenAI API Key not found in settings
Gemini API Key not found in settings
Solution: Verify the API key is set in Settings or .env file.

​
Rate Limiting

Both providers have their own rate limits:
  • OpenAI: Depends on your account tier
  • Gemini: Free tier has daily quotas
If you exceed provider limits, users will receive a generation failed error. Consider upgrading your API plan or implementing request queuing.

​
Model Availability

If a model becomes unavailable:
  1. Check provider status pages
  2. Update the model name in AiService.php
  3. Test with the alternative provider

​
Security Considerations

Never expose API keys in frontend code or commit them to version control.
  • API keys are stored in the settings table or .env
  • Keys are never sent to the client
  • All AI requests are server-side only
  • User prompts are not stored permanently

​
Next Steps

Automation

Schedule AI-generated messages for automatic delivery

Templates

Save AI-generated content as reusable templates

Build docs developers (and LLMs) love

Get started for freeTalk to us