LLM-OS: A conversational interface to your local OS

This agent, built with LangCode and LangGraph, can execute Python code in a stateful Jupyter Notebook with full access to your system, The Internet, and output images generated inside Jupyter environment (ex. Matplotlib plot image).

Installation

1. Clone the repository

git clone https://github.com/keell0renz/llm-os.git

2. Set up .env file

Please copy .env.example into .env and substitute placeholders with your data and your preference about model asking you every time it wants to execute code.

3. Install dependencies

poetry install

4. Run chainlit app

cd llm-os/

poetry run chainlit run llm_os/app.py