The title here is inaccurate: this post is about how I've vibe coded 124 things and most of them are non-useful... but a small subset of them are genuinely useful and worth highlighting.
swah · 9h ago
Remembering the tools when they are needed is hard, I guess? I have this issue with tools that I only used once[1] or one full day but go for weeks forgotten.
[1]Which now make sense to make since cost is almost zero...
Tarsul · 8h ago
I just checked his Hacker News Multi-Term Histogram[0] and I like it! E.g. if you compare Python to Java from 2015 on, you can clearly see that Java was up front in 2015 and since about 2023 Python is winning (although by less than I imagined). There are probably quite a few terms that might be interesting.
[0]https://tools.simonwillison.net/hacker-news-histogram
anotherhue · 15h ago
I think we're witnessing the evolution of the one-liner.
Most of these have extremely limited scope, and are probably not going to be invested in long term as software requires, they're throwaway.
Which is exactly what horrible find to xargs to inline perl was.
I personally feel less annoyed by this vibe mania when I think of them this way.
Note: two of them seem at face value to be equivalent to gpg or imagemagick CLI calls but hyper (over IMO) specialised.
Disposal8433 · 14h ago
I had the same reaction. It's a big stretch to call a "tool" 500 lines of JS that could be replaced with a few Bash calls, a regexp, or a small self-contained Python+UV script. Even worse when it's limited in functionality.
We failed at creating automation tools for the users (even if Apple tried that with AppleScript or Automator), but vibe coding throwaway scripts is not the answer.
simonw · 13h ago
I can't run a few bash calls or a Python+UV script in Mobile Safari on my phone. That's very much a goal of most of these as well.
TBH using a bull and a bear for up/down is stupid.
Why not giraffe and mouse or something clearer?
simonw · 13h ago
I should note that "124 useful
tools" is commentary in the Hacker News title submission here - the point of the linked post is that most of them are not useful so it highlights some of the ones that are.
That aid I've found the bullish-bearish one totally useful myself, it finally helped me remember which was which!
[1]Which now make sense to make since cost is almost zero...
Most of these have extremely limited scope, and are probably not going to be invested in long term as software requires, they're throwaway.
Which is exactly what horrible find to xargs to inline perl was.
I personally feel less annoyed by this vibe mania when I think of them this way.
Note: two of them seem at face value to be equivalent to gpg or imagemagick CLI calls but hyper (over IMO) specialised.
We failed at creating automation tools for the users (even if Apple tried that with AppleScript or Automator), but vibe coding throwaway scripts is not the answer.
I actually have a separate collection of vibe-coded UV+Python scripts on that site here: https://tools.simonwillison.net/python/
Hey LLM, give me some examples for X Y Z tools, and provide doc snippets for the command line parameters being used in the examples.
Rather than spending 2-3 hours scouring the Internet and having 50 tabs open, you can have it in minutes.
If you have to resort to this kind of hyperbole to make vibe-coding seem useful, maybe it isn't that useful?
Even better when you have them all in a repo w/ an agent like Codex or Claude Code to constantly tweak/remix them as needed.
https://tools.simonwillison.net/bullish-bearish
Come on..
Why not giraffe and mouse or something clearer?
That aid I've found the bullish-bearish one totally useful myself, it finally helped me remember which was which!