luxkun/ReGoap

Health v2 · maintenance only

Adoption not part of the score

1111 stars · 155 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

ReGoap is a generic C# GOAP (Goal Oriented Action Planning) library with adapters for Unity and Godot. It provides an engine-agnostic planner core that uses A* search to build action sequences for agents based on goals, actions, memory, and sensors.

Use cases

• implement NPC AI with goal-oriented action planning in Unity
• create emergent enemy behavior without hardcoded state machines
• build simulation agents that replan when world state changes
• define agent goals and actions with preconditions and effects in C#
• use GOAP planning in Godot games
• model inventory and resource-driven NPC decisions like crafting or gathering

When to choose

• you want flexible, emergent NPC behavior instead of behavior trees or switch-based AI
• you need a reusable, engine-agnostic GOAP core with Unity or Godot integration
• your agents must replan dynamically in response to world changes
• you need comparator-based conditions for counts, thresholds, and inventory logic

When to avoid

• you need simple scripted enemy behavior that a state machine handles fine
• you want a visual behavior tree editor or node-based AI authoring
• your project is not in C# or does not use Unity/Godot
• you need machine-learning-driven agents rather than symbolic planning

Facets

library · maturity active

game-engine simulation machine-learning artificial-intelligence simulation cross-platform game-engine goap goal-oriented-action-planning npc-ai unity godot csharp gamedev a-star-planning game-development

Member repositories

For agents

markdown · JSON · MCP: product_card(name="luxkun/ReGoap")