A flexible multi-agent system using the Langchain's agent SDK, built with Python and Ollama LLama models that enables autonomous agents to collaborate on complex tasks through sequential and parallel workflows as well as access to an MCP Server (Model Context Protocol).
This project implements an Agent-to-Agent (A2A) communication framework where multiple AI agents with specialized roles can work together to accomplish tasks, As of now this system can only work locally. Agents can operate:
- Sequentially – Passing outputs as inputs to the next agent.
- In parallel – Running multiple agents simultaneously and aggregating results. (Currently in Progress)
The system is designed to be modular, so new agents or tools can be easily added.
- Multi-agent orchestration with a triage agent directing tasks.
- Integration with Ollama LLaMA models for high-quality AI reasoning.
- Asynchronous execution using
asyncio. - Easily extensible with new specialist agents.
- Locally host MCP (Model Context Protocol) Server to access tools (Very basic toolset currently)
- Python 3.11 or higher
- Llama3.1 or higher
- Ollama CLI installed and configured
- Virtual environment recommended