AutoChat
- AutoChat watches incoming chat and keeps a rolling context history (configurable size).
- API key source:
WURST_OPENAI_KEYorWURST_DEEPSEEK_KEY(set in Java/Minecraft, not your PC).- Can also set in-game (Not recommended)
- Replies when directly addressed, in clear 1-on-1 chat, or freely based on configurable butt-in chance.
- Can prioritize the newest message while keeping older lines as secondary context.
- Mention detection supports your username plus nickname/derivative variants.
- Optional
Always reply when mentionedbypasses reply-gap timers for direct mentions. - Editable system prompt in-game.
- Persona is configurable via text field and overrides whatever is in the system prompt.
- Model is selectable, including:
chatgpt-4o-latest,gpt-5.2-chat-latest,gpt-5.2,gpt-4o,gpt-5.1-chat-latest,gpt-5-chat-latest,gpt-5.1,gpt-5,gpt-4.1,gpt-5-mini,gpt-4.1-mini,gpt-4o-mini,gpt-4.1-nano,gpt-5-nano. - Added DeepSeek support as an additional provider/model option.
- Models are now filtered by provider, with separate API keys for OpenAI and DeepSeek.
- DeepSeek-specific reasoning/thinking controls are available when using that provider.
- Usage tracking can show token and cost totals in the HackList.
- Generation controls: max tokens, temperature, max output chars.
- Timing controls: min reply gap, min unsolicited gap, optional queued scheduling for gap-blocked replies, WPM pacing (toggle + words-per-minute), and max concurrent requests.
- Debug mode can print raw OpenAI request/response JSON in the console.
- Safety/sanitization: filters likely prompt-injection attempts, strips leading
<Name>/[Name]style prefixes, prevents slash-command output, enforces single-line replies, and normalizes output to ASCII-safe chat text.
Parent: What's new in this fork?