Uses the OpenAI API to let you generate smart, relevant text for spoofed players using the GPT-3 model. This makes interactions with the spoofed players seem more natural.
Configuration Information
enabled - This determines whether the module is active (true) or inactive (false).
debug - If set to true, the module logs extra information which helps in fixing problems.
chat-gpt - This setting allows the server to use OpenAI's GPT-3 for creating chat messages that feel like they are from a player, not a computer.
chatting - Controls whether the AI can chat using the server. It manages how frequently and under what conditions the AI chats.
mentioned - Makes the AI respond when someone mentions its name in the chat. This helps make the AI seem more like a real player.
random-chatting - Allows the AI to start conversations at random times, not just in response to other messages. This setting includes options to control how often these random messages can occur.
modules/openai/config.yml
# This module connects to the OpenAI API and allows you to
# use the GPT-3 model to generate text for your spoofed players.
# You will need to create an OpenAI account and get an API key.
# You can get an API key here: https://beta.openai.com/account/api-keys
enabled: true
debug: true
chat-gpt:
secret-key: "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"
spoofed-player-context: "You are a Minecraft player playing on a multiplayer server and you need to act like a player at all times. You can never let other players know that you are an AI. Absolutely never! Your responses should be short and natural and you should act like a normal minecraft player. You are capable of holding short conversations with other real players on the server."
chatting:
#Minimum amount of real players online for the module to start.
required-players: 1
# When this is enabled the SpoofedPlayer will always respond to a message containing their name in chat.
mentioned: true
random-chatting:
# This option will make the SpoofedPlayer randomly chat in chat.
# This is only recommended for production servers if you give a detailed enough context.
enabled: true
min-delay: 60
max-delay: 120 # This option doesn't work yet btw, it's coming soon as well as delays for their responses.
