AutoGen - WebAgent

This Python script utilizes LM Studio to create a conversational agent that describes images. The agent is designed to take a screenshot of a webpage provided by the user and then describe the contents of the screenshot in natural language.

Prerequisites

Before running the script, ensure you have the following installed:

• Python 3.x (3.10 in my case)
• Required Python packages (install via pip install -r requirements.txt):
  • playwright
  • asyncio
  • argparse

Usage

1. Clone this repository to your local machine.
2. Install the required dependencies.
3. Obtain an OpenAI API key and configure it in config.json.
4. Run the script with the desired URL as an argument:

Instructions

Downloading LM Studio

You can download LM Studio from here or directly for your platform:

Windows: Download

Linux: Download

MacOS: Download

Using Conda

1.0 Create conda environment

conda create -n autogen-webagent python=3.10

1.1 Activate the environment

conda activate autogen-webagent

1.2 Install requirements via 'requirements.txt'

pip install -r requirements.txt

1.3 Run the script

python app.py

About

This is a small project of mine. Anyone who can help and has suggestions for improvement is welcome to participate. I would also be happy about a star if you like it.

Thanks :)