MUXI Framework Documentation

Welcome to the official documentation for MUXI, an extensible framework for building AI agents with real-time communication capabilities, memory persistence, and MCP server integration.

This project is a work in progress and is not yet ready for production use. The framework is actively being developed with new features being added regularly. Please refer to the roadmap for information about the current state of the project and where it’s headed.

Documentation Structure

Our documentation is organized to serve both newcomers and experienced developers:

Introduction

Start here to understand the MUXI Framework, its core concepts, and how to get started:

• Overview & Key Concepts
• Why MUXI?
• Quick Start Guide
• Architecture
• Installation

Building Agents

Learn how to create and configure agents for different purposes:

• Simple Agents
• Multi-Agent Systems
• Adding Memory
• Agent Configuration

Extending Capabilities

Enhance your agents with external services and context knowledge:

• Using MCP Servers
• Creating Custom MCP Servers
• Context Knowledge Integration
• Multi-Modal Support

Interfaces & Clients

Discover the different ways to interact with MUXI:

• Command-Line Interface
• Server Deployment
• Web Dashboard
• WebSocket Communication

Technical Deep Dives

Detailed information for developers who want to understand the framework at a deeper level:

• Agents & Models
• Memory System
• MCP System
• Communication

Reference

Detailed reference material:

• API Documentation
• Configuration Options
• Package Structure
• Examples Library

Development

Information for contributors and developers:

• Contributing Guidelines
• Testing Strategies
• Codebase Organization
• Changelog

Key Features

• Multi-Agent Orchestration: Create and manage multiple AI agents with different capabilities
• Intelligent Message Routing: Automatically select the most appropriate agent based on message content
• Model Context Protocol (MCP): Connect to external services via standardized MCP servers
• Memory Systems: Short-term buffer memory and long-term persistent memory
• Multi-User Support: Memobase provides user-specific memory partitioning
• Context Knowledge: Store and retrieve structured information to personalize agent responses
• Real-Time Communication: WebSocket support for streaming responses
• Flexible Configuration: Define agents using YAML or JSON with minimal code

Getting Started

The fastest way to get started with MUXI is to follow our Quick Start Guide.

Community & Support

• GitHub Repository
• Issue Tracker

License

This project is licensed under a dual licensing model to balance open-source collaboration with sustainable business practices.

During the development phase, the software is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 (CC BY-NC-ND 4.0) license.