config.yml

Global settings are created at plugins/Blockfolk/config.yml. Restart the server after changing provider settings.

General settings

Key	Default	Description
chat-input-timeout-seconds	60	Time allowed to answer a Blockfolk admin chat prompt.
question-timeout-seconds	30	Time allowed to answer an NPC question before its cancel/timeout branch runs.
proximity-transition-cooldown-seconds	3	Debounces rapid approach/leave transitions while an NPC starts or stops moving.
mineskin-api-key	empty	Optional key for higher MineSkin limits when resolving external skin image URLs.

OpenRouter settings

Key	Default	Description
openrouter.endpoint	https://openrouter.ai/api/v1/chat/completions	Chat-completions endpoint. HTTPS is required.
openrouter.api-key	empty	OpenRouter API key. Leave empty to keep AI unavailable.
openrouter.model	deepseek/deepseek-v4-flash-0731	OpenRouter model identifier.
openrouter.timeout-seconds	12	Network request timeout. Increase it for slower providers.
openrouter.max-tokens	1600	Maximum output allowance for the final JSON decision.

The plugin requests JSON output with temperature 0.4 and disables model reasoning for lower gameplay latency.

AI control

Key	Default	Description
ai-control.invocation-cooldown-seconds	2	Per-NPC protection in addition to event-specific trigger throttling.
ai-control.conversation-history-limit	20	Recent conversation lines retained per NPC conversation. 0 disables history.

NPC-specific context, memory, inventory access, chat response, and allowed actions are configured through each preset's AI Behaviour menu, not in this file.

Complete default file

chat-input-timeout-seconds: 60
question-timeout-seconds: 30
proximity-transition-cooldown-seconds: 3
mineskin-api-key: ""

openrouter:
  endpoint: "https://openrouter.ai/api/v1/chat/completions"
  api-key: ""
  model: "deepseek/deepseek-v4-flash-0731"
  timeout-seconds: 12
  max-tokens: 1600

ai-control:
  invocation-cooldown-seconds: 2
  conversation-history-limit: 20