HACKER Q&A
📣 shaburn

Why aren't we using ChatGPT in the CLI?


Why aren't we using ChatGPT in the CLI?


  👤 jasonjmcghee Accepted Answer ✓
Many people are. It's quite popular to build yourself as it's quite easy. There are also more extensive implementations like https://github.com/KillianLucas/open-interpreter which is wildly popular (>27k stars).

You can also use local LLMs via cli which is also quite popular https://github.com/jmorganca/ollama


👤 arthurcolle
you aren't? There are a bunch

https://i.imgur.com/ZH9ZPr2.png

Open Interpreter is really good


👤 muzani
Made one myself, then found this on HN and have been using it since: https://github.com/npiv/chatblade

Some say it's because it's not free, but it just feels good


👤 swah
I use a random one [1], when I remember. Type what I need in english and press Ctrl-G:

    ~/ git clean branch⌛ (ctrl-g gives me the little hourglass)
wait for a second and the command line is replaced:

    ~/ git clean -f -d && git checkout branch
Of course when the result is not what you want, might as well open the browser...

[1] https://github.com/TheR1D/shell_gpt


👤 logicalmonster
While clever as a timesaving tool to look up syntax, it seems kind of dangerous to rely on AI for this without checking and verifying its suggestions. CLIs are very powerful. You can make big changes to complex systems with one random one-liner with the right flags.

Now the problem is that state of the art AI can seem incredibly smart in a large portion of cases, but it might act jaw-droppingly stupid in that 1% of cases. You can get used to trusting it too much and accidentally make a bad command. If you're trusting that the machine's suggestions are right and aren't careful, you can easily blow up your system.

Do you trust AI to not royally screw up your system that 1% of the time it makes a goofy suggestion?