A Python-based Azure CLI assistant that provides natural language processing capabilities for Azure commands, leveraging Azure MCP Server.
- Python 3.11+ (as specified in pyproject.toml)
- Azure CLI installed and configured
- Ollama with a model downloaded
- uv for Python package management
- Azure MCP server installed and running
-
Clone the repo
git clone https://github.com/yourusername/azure-terminal-copilot.git cd azure-terminal-copilot
-
Open the terminal and Start a virtual env with uv
uv venv
-
Install packages using uv
uv pip install .
-
Run Ollama and make note of its local address
-
Run Azure MCP server and make note of its local address
-
Rename
.env-sample
to.env
-
I provided dummy values there so make sure to update with the values that correspond to your locally running ollama, Azure MCP, and model you want to use
-
Now you can run
python main.py
Once the program is running, try a few things:
- Try providing a query like 'list all my resource groups', is the command you expect to be executed being ran?
- Try different models, notice which ones perform better?
- Try tweaking the system prompt, how would you improve it?
- Make sure your azure cli is logged in, azure MCP uses that as auth
This project is licensed under the MIT License - see the LICENSE file for details.