I have forked this project because it's very in-line with what I've been trying to do. If you're reading this, you're probably in the same boat.
I'm aiming at making a privately hosted, personal AI with the user's data privacy being one of the top concerns. Another major concern is ease of use for all users. No joke: I want my parents to be able to pick up a device with an AI and not have to perform any tech support or explain anything to them. More details are in development.
- Windows Copilot is coming out.
- Linux users don't have AI integration.
- Free software forever!
It is structured in 2 parts, the client and server. Why this approach? I chose it because we'll probably be using multiple different UIs for the assistant depending on the distro/DE. Therefore it makes sense to have a server rather than implementing the whole project from scratch in a different UI.
If you'd like to contribute we are accepting input and changes in any of the source code. you can raise an issue or create a pull request. thanks :)
- if you're a prompt engineer feel free to take a look at System.MD (the system prompt)
- If you're a Python developer feel free to take a look at server.py (the heart of this project.)
- Also look at fiosa.py but it is obsolete.
(!) = priority
- Running commands
- Implement server and basic CLI client
- Create clients for desktop environments like GNOME (!)
- Look into using Microsoft Guidance framework to stop user perspective hallucinations
- create early builds of Linux distro