So I stumbled upon this interesting project on GitHub the other day. It’s called Skill, and the idea is pretty straightforward: you type (or speak) a natural language command, and it executes actions on your computer.
No more digging through menus, remembering keyboard shortcuts, or writing complex scripts. Just say something like “open Chrome and start a new incognito window” or “mute Spotify and pause the video” — and it just works.
The project is still fresh, but the architecture is already solid. It uses a plugin-based system, so you can add support for more apps as you go. Out of the box, it supports browser control, media players, system commands, and even some IDE actions. The setup is dead simple: one command to install, then you define your triggers.
What I really like is the privacy-first approach. Everything runs locally — no cloud dependency, no data leaving your machine. It uses local NLP models for understanding, so you don’t need to worry about your commands being sent somewhere.
For developers, it’s especially handy. Want to quickly run a terminal command without switching windows? Just type “run npm install in the current project folder.” Or “open the terminal and show the last 10 lines of the log file.” It’s like having a personal assistant that actually understands your workflow.
The project is MIT licensed, so you can fork it, customize it, or even embed it into your own tools. If you’re tired of repetitive clicks and want a more fluid way to interact with your computer, this is definitely worth a look.
If you’ve got a few minutes, give it a spin. It might just change how you think about controlling your desktop.