So far I’ve created a flask server that uses two models. One is a reasoning model QWEN3 and the other one is a vision model. My AI can read documents, analyze your screen run power shelf commands, and I’m looking to extend the automation even further I want to add in GUI interaction so essentially I would talk to my computer and it would do the tax I wanted to do for instance chrome go to youtube.com search for a certain video and play it I’m trying to create AI system that exists on top of my system that can control the computer via my voice there any repositories that I could use keep in mind I want to make this local only