This is an initial version of a chat support chat bot that guides you on how to do different actions across websites and other software
The bot uses MultiModal abilities of LLMs to capture the screen and than reasons on what should be the next steps
pip install -r requirements.txt
- Put your openAI API key in the relevant line in the code
- Run the code
- Add support for Anthropic Claude3
- Add support for open-source MultiModal LLMs (LLava, Baqllava, etc)
- Improve chat history
- Add the ability to upload technical guides to help the bot give better answers
- Add a caching mechanism based of VectorDB to reduce cost and improve latency
Contributions are welcome. Please open an issue to discuss the changes you would like to make.