🧠 LLM-Powered Terminal Assistant using A2A & MCP

youtube link : Deep dive and step by step tutorial of A2A and MCP, including code walthrough

This is an LLM-powered application that allows users to execute terminal commands using simple English queries like:

“Delete readme.txt on my desktop.”

It leverages the A2A (Agent-to-Agent) and MCP (Model Context Protocol) to interpret and execute commands on a Linux system.

---

🛠️ Getting Started

Follow these steps to get the system up and running.

1. Clone the Repository

git clone https://github.com/ishanExtreme/a2a_mcp-example.git
cd a2a_mcp-example

2. Set OpenAI API Key

Export your OpenAI API key to your environment:

export OPENAI_API_KEY=""

3. Create and Activate Virtual Environment

python -m venv venv or conda create ...
source venv/bin/activate  # On Windows use: venv\Scripts\activate

4. Install Requirements

pip install -r requirements.txt

---

🧩 Running the Application

Ensure all the services are running in separate terminal tabs/windows:

🖥️ Start the UI Server

cd ./client/demo/ui
uv run main.py

🔁 Start the A2A Server

cd ./a2a_server
python main.py

🖥️ Start the MCP Server

cd ./mcp_server
python mcp_server.py

---

🌐 Accessing the Application

1. Open your browser and go to: http://localhost:12000
2. Navigate to the Agents section.
3. Add a Linux agent using the following endpoint:

http://localhost:10000

You are now ready to start chatting and executing Linux commands via natural language!

---

🧪 Example Queries

• “Delete the notes.txt file on Desktop”
• “List all files in the Documents folder”
• “Create a new directory called projects”

---

🧠 Built With

• OpenAI GPT
• A2A Protocol
• MCP Protocol
• FastAPI + Uvicorn
• Python 3.12+

---

📜 License

MIT License © 2025

---