Digimon Engine

Digimon Engine is an open - source multi - agent, multi - player game framework focusing on AI - native game and intelligent metaverse development. It supports social and financial AI agents, provides an immersive game experience, and is committed to building an AI agent environment similar to that in 'Westworld'.

Games and gamification Developer tools #AI game #Multi - agent #Open - source framework #Real - time communication Local .TypeScript

rating : 2.5 points

downloads : 6.2K

update time : 2025-04-29

Open Site

What is Digimon Engine?

Digimon Engine is an AI game development platform similar to Unreal Engine. It supports the construction of AI agents with social and financial functions, creating an immersive AI - native game experience. It aims to build an agent environment similar to that in 'Westworld'.

How to use Digimon Engine?

Developers can build games by integrating the MCP protocol, DAMN.FUN SDK, and the Digimon Engine architecture, supporting seamless connection of external clients, LLMs, and AI agents.

Applicable scenarios

Suitable for building AI - driven metaverse games, social simulation environments, and virtual worlds that require a large number of agent interactions.

Main features

Multi - agent system

Each agent has a unique identity and motivation, and can move freely, talk, and build relationships in the world.

Memory system

Agents can store and retrieve past interaction memories, and achieve persistent memory through a vector database (Pinecone).

Event scheduling

The engine automatically schedules agent activities, predicts and handles interaction collisions between agents.

Low - latency design

Adopt the compressed state (HistoryObject) technology, updating 60 times per second to ensure a smooth game experience.

Advantages

A completely open - source AI game engine

AI agents are treated equally with human players, no second - class NPCs

Built - in memory and social systems to create more realistic interactions

Designed specifically for AI - native games with performance optimization

Limitations

Requires certain TypeScript development experience

The current community and ecosystem are still in development

Documentation and tutorial resources are relatively limited

How to use

Environment preparation

Install Node.js 23.3.0 and npm 11.0.0

Clone the repository

Get the Digimon Engine source code

Run the example game

Start the built - in DAMN game example

Usage cases

DAMN game

The first AI - native game built using Digimon Engine

Agent social simulation

Create AI characters with memory and social abilities

Frequently Asked Questions

What types of game development is Digimon Engine suitable for?

Is game engine development experience required?

How to get technical support?

Related resources

Official documentation

Complete documentation for Digimon Engine

GitHub repository

Source code and issue tracking

Telegram community

Developer communication group

DAMN game demo

An example game built using Digimon Engine

🚀 Digimon Engine

Digimon Engine is an AI-native multiplayer online game framework designed to build dynamic, intelligent, and immersive game worlds, aiming to create the next-generation metaverse-level gaming experience.

🚀 Quick Start

Prerequisites

Node.js (Recommended version: 23.3.0)
npm (Recommended version: 11.0.0)

Community and Contact

GitHub Issues: Report problems and propose feature requests.
Telegram: Join the community discussion and share your works.

✨ Features

Core Features of Digimon Engine

Equal Status of AI and Humans: Place AI agents and human players in the same operating cycle, giving AI equal decision-making power.
Low Latency and High Scalability: Support more players and agents while maintaining smoothness through innovative state compression technology.
Modular Architecture: The componentized design allows developers to flexibly replace or expand various functional modules.

Key Features of MCP Server

The MCP (Multi-Proxy) server is the core service of Digimon Engine, responsible for handling game world state updates, player and agent interactions, and providing the following key functions:

External Client Integration: Support seamless docking with existing game engines and frameworks.
Integration of LLMs and AI Agents: Allow the use of large language models and other AI technologies to enhance game intelligence.
Dynamic Rule Execution: Adjust game logic based on real-time data to ensure fairness and fun.

📚 Documentation

Project Links

Project Overview

Digimon Engine is an AI-native multiplayer online game framework designed for building dynamic, intelligent, and immersive game worlds. It supports the interaction of multiple agents (AI) and players in the same environment, aiming to create the next-generation metaverse-level gaming experience.

MCP Server Overview

Functional Description

The MCP (Multi-Proxy) server is the core service of Digimon Engine, responsible for handling game world state updates, player and agent interactions, and providing the following key functions:

External Client Integration: Support seamless docking with existing game engines and frameworks.
Integration of LLMs and AI Agents: Allow the use of large language models and other AI technologies to enhance game intelligence.
Dynamic Rule Execution: Adjust game logic based on real-time data to ensure fairness and fun.

Architectural Components

State Manager: Responsible for maintaining the current state of the game world and processing inputs from players and agents.
Event Log: Record all events (such as player movement, item picking up, etc.) for playback and analysis.
Rule Engine: Define and execute the logical rules of the game to ensure that the behavior of the game world complies with the design specifications.

Overall Architecture Overview

Core Components

Scheduler: Periodically trigger new simulation steps to coordinate the work rhythm of each module.
Game Engine:
- Responsible for loading and updating game data.
- Handle the actions of all participants (players and agents).
- Calculate and save state changes to the database.
Event Logger: Record a snapshot of the game world at each timestamp to support the playback function.
Memory and Vector Database:
- Store the historical behavior data of AI agents for subsequent decision-making reference.
- Use a vector database for efficient knowledge retrieval and association.

Design Philosophy

Unified Loop Mechanism: All actions, whether from players or agents, are processed in the same time loop to ensure fairness and consistency.
State Compression Technology:
- Save a complete snapshot every second and record key changes (such as position) every frame.
- Clients can obtain complete snapshots and historical data on demand to achieve a low-latency and smooth experience.

Core Challenges and Solutions

Latency Optimization: Significantly reduce network transmission overhead through state compression and hierarchical storage.
Scalability Design:
- Support horizontal expansion to easily handle the access of more players and agents.
- Adopt a distributed architecture to ensure high availability and load balancing.

Release of the First AI-Native Game

The First Game: DAMN (https://damn.fun)

Introduction: An exciting AI-native multiplayer online game.
Features:
- Truly achieve equal interaction between AI and players,告别传统的NPC脚本。
- Dynamically generate the game world to provide an infinite number of possible gameplay experiences.