Bunnymanbrain

Based on game build 7140014 | Last updated: 2026-03-03

Overview

BunnymanBrain is a behavior tree-based AI controller for bunnyman entities. It implements a priority-based decision system that handles fleeing (from threats, fire, acid rain, low health), pursuing and attacking targets (including scarers), seeking food, trading with players, following a leader, and returning to a home when appropriate. It relies heavily on common DST components like health, combat, eater, homeseeker, follower, and trader, and uses a prioritized behavior tree to coordinate actions.

Usage example

local inst = CreateEntity()
inst:AddTag("bunnyman")
inst:AddComponent("health")
inst:AddComponent("combat")
inst:AddComponent("eater")
inst:AddComponent("homeseeker")
inst:AddComponent("follower")
inst:AddComponent("trader")
inst:AddComponent("hauntable")
inst:AddComponent("burnable")
-- ... (initialize home and leader if needed)
inst.brain = BunnymanBrain(inst)
inst.brain:OnStart()

Dependencies & tags

Components used: health, combat, eater, edible, equippable, follower, trader, hauntable, burnable, inventory, inventoryitem, homeseeker, playercontroller.
Tags: Checks "manrabbitscarer", "INLIMBO", "outofreach", "NOCLICK", "burnt"; adds no tags itself.

Properties

Property	Type	Default Value	Description
inst	Entity	assigned at construction	Reference to the entity instance the brain controls.
bt	BehaviorTree	nil (set in OnStart)	The root behavior tree, initialized and assigned during OnStart.

Main functions

OnStart()

• Description: Constructs and initializes the behavior tree using a priority-ordered sequence of conditionally activated nodes. It sets up a behavior tree that evaluates high-priority threats (panic triggers) first, then falls back to combat, trading, foraging, following, home-seeking, and wandering.
• Parameters: None.
• Returns: Nothing.
• Error states: Requires that inst.components has all necessary components (health, combat, eater, homeseeker, follower, trader, hauntable, burnable, inventory, inventoryitem) attached before calling; otherwise node execution may fail.

Events & listeners

• Listens to: None — this brain uses a reactive behavior tree evaluated each tick rather than event-driven scheduling.
• Pushes: None — the brain itself does not fire events, though its sub-behaviours (e.g., Panic, RunAway) may.