Yolo: Natural Language to Shell Commands with ChatGPT API
Once in a while I go build some fun new tools to adopt new tech. Just last week OpenAI made their gpt-3.5-turbo
model accessible via API endpoints.
Update: The latest version also supports GPT-4.
So, I thought it’s time to start building a tool to leverage it.
What is yolo?
Do you know those moments when you can’t remember a shell command, or some arguments to it? How do you pipe all errors to /dev/null again? Things along those lines. This is where yolo comes to the rescue.
You live only once!
Yolo translates your question to a shell command (like bash
, zsh
or PowerShell
on Windows), and then executes the command. By default, it won’t right away execute the command (but that can be overwritten, if you desire the true yolo-style).
This project was created for fun, but it’s actually kind of useful.
I started building yolo some time ago, and now upgraded it from the GPT-3 completion API to the brand new ChatCompletion (ChatGPT) API.
How to install and use yourself
The code is here yolo-ai-cmdbot Github repo.
Easy install, don’t forget to set your OpenAI API key.
I mostly use it on Linux and some macOS, Windows is less tested.
Detailed walk-through and demonstration
There is also a YouTube Video that shows it in action:
Happy hacking!
References
- Github Repo for yolo: https://github.com/wunderwuzzi23/yolo-ai-cmdbot
- ChatCompleation API documentation: https://platform.openai.com/docs/guides/chat