ActGPT is a chatbot designed to automate web browser interactions based on natural language instructions. It targets users who need to perform repetitive web tasks, such as data scraping, social media posting, or code generation, by translating high-level commands into browser actions.

How It Works

ActGPT leverages a large language model (LLM) to interpret user prompts and generate a sequence of browser automation steps. It uses Selenium WebDriver to control a Chrome browser instance, allowing it to navigate websites, locate elements, input text, and click buttons. The system's advantage lies in its ability to translate abstract requests into concrete, executable browser actions, simplifying complex workflows.

Quick Start & Requirements

• Install dependencies: pip install -r requirements.txt
• Run chatbot: python3 demo.py
• Prerequisites: OpenAI API key, chromedriver executable, Python 3.x.
• Configuration: conf/config.yaml for API key and chromedriver path. User data directory for browser cookies can also be specified.

Highlighted Details

• Automates tasks like Wikipedia searches, Twitter posting, and code generation.
• Supports multi-step instructions, including conditional logic and variable passing.
• Can interact with web elements based on text content, visibility, and proximity.
• Demonstrates capabilities for data extraction, summarization, and AI-driven content creation.

Maintenance & Community

The project appears to be a personal project by ethanhe42, with no explicit mention of community channels, roadmap, or significant contributor activity in the provided README.

Licensing & Compatibility

The README does not specify a license. Compatibility for commercial use or closed-source linking is undetermined.

Limitations & Caveats

The project is presented as a demonstration, and its robustness for complex or production environments is not detailed. Error handling, advanced browser automation features, and support for browsers other than Chrome are not specified.