AI

The ai.toml file configures the AI chat functionality of the bot.

Template

activate_ai          = false
 
provider             = "groq" # Available providers: "ollama", "gemini", "groq"
model                = "qwen/qwen3-32b"
max_messages_context = 10
remove_emojis        = true
lower_response       = true
answer_to_reply      = false
include_username     = true
include_reply        = true
include_attachment   = true
 
system_prompt = """
You are a helpful assistant that gives short, helpful answers.
Your answers can maximally be 1000 characters long.
"""
 
user_prompt = """
{{USERNAME}} wrote the following message: {{MESSAGE}}
"""
 
user_reply_prompt = """
They added the following context: {{REPLY}}
"""
 
user_attachement_prompt = """
The user attached the following file: {{FILE}}
"""
 
user_prompt_structure = """
{{PROMPT}}
 
{{REPLY_PROMPT}}
 
{{ATTACHMENT_PROMPT}}
"""

Values

Basic Settings

Value	Type	Description
activate_ai	boolean	Enable or disable AI chat functionality
provider	string	AI provider to use: "ollama", "gemini", or "groq"
model	string	Model to use (varies by provider)
max_messages_context	integer	Number of previous messages to include in context

Response Settings

Value	Type	Description
remove_emojis	boolean	Remove emojis from AI responses
lower_response	boolean	Convert AI responses to lowercase
answer_to_reply	boolean	Also answer when replying to the bot’s own messages

Context Settings

Value	Type	Description
include_username	boolean	Include the username in the prompt
include_reply	boolean	Include replied message content
include_attachment	boolean	Include attachment filenames

Prompt Templates

Value	Description
system_prompt	Instructions for the AI behavior
user_prompt	Template for user messages ({{USERNAME}}, {{MESSAGE}})
user_reply_prompt	Template for replied content ({{REPLY}})
user_attachement_prompt	Template for attachments ({{FILE}})
user_prompt_structure	How to combine all prompt parts

Provider Configuration

Groq (Default)

provider = "groq"
model    = "qwen/qwen3-32b"

Requires GROQ API key in .env file.

Gemini

provider = "gemini"
model    = "gemini-2.0-flash"

Requires GEMINI API key in .env file.

Ollama

provider = "ollama"
model    = "llama3"

Requires Ollama running locally (default: http://localhost:11434).

Examples

Enable AI with Groq

activate_ai = true
provider    = "groq"
model       = "qwen/qwen3-32b"

Custom system prompt

system_prompt = """
You are a sarcastic assistant who answers questions concisely.
Keep responses under 500 characters.
Never use emojis in your responses.
"""

Include more context

max_messages_context = 20
include_attachment  = true

Environment Variables

Make sure to set the appropriate API key in your .env file:

BOT_TOKEN=your_discord_bot_token
GEMINI=your_gemini_api_key      # Only if using Gemini
GROQ=your_groq_api_key         # Only if using Groq