What is Distri

Distri is a framework for building and composing AI agents, written in Rust. It allows developers to create, publish, and combine agent capabilities using the MCP (Multi-Agent Communication Protocol) standard.

Use cases

Use cases for Distri include creating AI agents for data exploration, automating workflows, and developing applications that require collaboration between different AI capabilities.

How to use

Distri can be used by configuring agents through YAML files or Rust scripts. Users can run commands to list agents and execute specific agents using sample configuration files.

Key features

Key features of Distri include a modular design for building independent agent modules, standardized communication through the MCP protocol, high performance due to Rust, the ability to publish tools as MCP-compatible agents, and easy integration with a simple API.

Where to use

Distri can be utilized in various fields such as AI development, automation, and collaborative task execution, where multiple agents need to communicate and work together.

Distri: A Composable Agent Framework

Distri is a framework for building and composing AI agents, written in Rust. It enables developers to create, publish, and combine agent capabilities using the MCP (Multi-Agent Communication Protocol) standard.

Getting Started

Distri agents can be configured and run in two ways:

1. YAML Configuration

2. Rust Scripts (Advanced Workflows) Coming Soon

Lets explore running distri using a sample configuration file.

List Agents

distri list -c samples/config.yaml

You can run github_explorer using:

distri run -c samples/config.yaml github_explorer

Installation

You can install Distri in two ways:

Using Cargo

cargo install --git https://github.com/distrihub/distri distri --locked

Using Docker

docker run -it distrihub/distri

MCP Proxy & Tools

Distri proxy also provides a convenient proxy to run stdio commands.

distri proxy -c samples/proxy.yaml

For looking at all the available tools

distri list-tools -c samples/config.yaml

AI Gateway

Distri is connected to AI Gateway and has access to 250+ LLMs. For more details checkout Langdb AI Gateway.

What is MCP?

MCP (Multi-Agent Communication Protocol) is a standardized protocol that enables agents to:

• Communicate with each other in a structured way
• Share capabilities and tools
• Execute tasks collaboratively
• Handle state management and coordination

With MCP, any agent can be published as a reusable tool that other agents can leverage, creating an ecosystem of composable AI capabilities.

Features

• 🔧 Modular Design: Build agents as independent modules that can be mixed and matched
• 🤝 MCP Protocol: Standardized communication between agents
• 🚀 Rust Performance: Built with Rust for reliability and speed
• 📦 Tool Publishing: Share your agents as MCP-compatible tools
• 🔌 Easy Integration: Simple API for adding new capabilities

Configuration

Distri agents can be configured in two ways:

1. YAML Configuration

2. Rust Scripts (Advanced Workflows) Coming Soon

Status

⚠️ Early Development: Distri is in early stages of development. APIs and protocols may change as we gather feedback and improve the framework.

Getting Started

[Documentation and examples coming soon]

License

This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

Contributing

We welcome contributions! Please see our CONTRIBUTING.md guide for details.