Overview
OpenGameAgent is a TypeScript runtime for AI-native game characters and worlds. It hosts a message-or-tool Agent loop while the game remains responsible for rules, permissions, authoritative state, and world writes. It is an MIT-licensed source preview under active development.
The runtime includes a Pi-backed Agent loop, exact run control, a single SQLite authority, reliable action exchange, and three thin engine integrations for Unity, Godot, and Unreal Engine. It also provides optional structured memory, durable provider-response suspension, durable goals/signals/schedules/mailboxes, structured player interactions, evidence-validated behavior Skill evolution, and more.
The architecture is composed of many packages like @opengameagent/protocol, @opengameagent/agent, @opengameagent/characters, etc. It supports Unity 6, Godot, and Unreal Engine through C# and C++ integrations. There is one Agent loop; optional memory, planning, scheduling, actions, media, realtime, integrations, and engine adapters compose around it rather than replace it.
Pricing
Free
Detailed plans have not been confirmed in our catalog. Check the official website for current limits and billing terms.
Visit WebsitePrices and limits may change. Confirm the currency, billing period, seat minimum and usage caps on the official website.
Use Cases
- Teams use OpenGameAgent to integrate AI-native characters into Unity games.
- Teams use OpenGameAgent to add AI characters to Godot games.
- Teams use OpenGameAgent to add AI characters to Unreal Engine games.
- Teams use OpenGameAgent to manage durable game state with SQLite.
- Teams use OpenGameAgent to implement structured memory and recall.
- Teams use OpenGameAgent to handle real-time speech interactions.
Features
C# runtime for game agents
Structured context for AI NPCs
ReAct-based tool use
Complex task planning
Agent memory support
Reliable in-game actions
Works with game engines and servers
Reviews
No reviews yet. Be the first to share your experience!