Agentum logo

Agentum

by BulloRosso

Agentum is a scalable, multi-component system designed for agent-to-agent (A2A) communication and interaction with large language models (LLMs) through the Anthropic Model Context Protocol (MCP). It provides a comprehensive framework for building and deploying AI agents with tools, resources, and real-time updates.

View on GitHub

Last updated: N/A

What is Agentum?

Agentum is a multi-component system that integrates an A2A server (following the Google A2A protocol) and an MCP server (Anthropic Model Context Protocol) with a React frontend, FastAPI backend, and Node.js proxy. It facilitates agent-to-agent communication, provides tools and resources for agents, and enables real-time updates.

How to use Agentum?

To use Agentum, you need to set up each component separately following the instructions in the README. This involves installing dependencies, running the backend API, frontend development server, proxy server, A2A server, and MCP server. Once all components are running, you can interact with the system through the frontend or directly via the API endpoints.

Key features of Agentum

  • Agent-to-Agent communication via A2A protocol

  • Integration with Anthropic's Model Context Protocol (MCP)

  • Real-time updates using Server-Sent Events (SSE)

  • Modular architecture with decoupled components

  • API documentation with Redoc

Use cases of Agentum

  • Building collaborative AI agents

  • Creating AI-powered tools and services

  • Developing LLM-based applications with access to external resources

  • Implementing agent-based task management systems

FAQ from Agentum

What is the A2A protocol?

The A2A (Agent-to-Agent) protocol is a communication standard that enables AI agents to interact and collaborate with each other.

What is the MCP server?

The MCP (Model Context Protocol) server provides tools, resources, and prompts for agents interacting with large language models.

What are the prerequisites for running Agentum?

You need Node.js 20+, Python 3.11+, and npm or yarn installed on your system.

How do I run the tests?

Navigate to the 'tests' directory and run 'node run-tests.js' to execute the integration tests for the A2A server.

Where can I find the API documentation?

The Backend API documentation is available at the /api/doc endpoint.