Here's an idea.

There have been over a dozen commits in the last 2 hours. If you are actually "vibe coding" a Prolog, then post the prompts you have used in the last 2 hours here.

If it helps to have a specific starting point, commit 7bbe652[0] is as good as any.

0 - https://github.com/nlothian/Vibe-Prolog/commit/7bbe652eaf0b0...

Wait, you think I'm coding this by hand! That is amazing. I love it!

I don't remember when which part happened, but I've been gradually moving over to the tools in https://github.com/nlothian/Vibe-Prolog/tree/main/ai-tools for most of the development. Most of the coding itself there is done with kilocode.

Most of the early conversations don't seem to be available in Claude code now (is there a way to see complete history?), but here's a typical session:

> can this interpreter pass command line arguments to the prolog program?

[no]

> add these as missing to @FEATURES.md

[done]

> use `gh issue` to create an issue to address this.

> - Explain comprehensively what needs to be done > - Note that new test cases must be created > - Note that @FEATURES.md needs updating

> Do not provide estimates

That raised: https://github.com/nlothian/Vibe-Prolog/issues/75

Here's the PR for that: https://github.com/nlothian/Vibe-Prolog/pull/92/files

That PR was generated using my fix-issue-with-kilocode command. The source code for that is all in here: https://github.com/nlothian/Vibe-Prolog/tree/main/ai-tools.

The code reviews were done with Codex and Gemini, and responded to using my address_pr_comments command in that same directory.

What are the odds :) I made this two weeks ago, also with Claude.

This one's a straight port, paip-python's[0] Prolog interpreter. (Itself based on Norvig's Paradigms of Artificial Intelligence Programming book, originally in Lisp.)[1]

https://github.com/a-n-d-a-i/prolog.js

It was in the web ui, so I expected it to give me a code block, but it spun up a vm, set up a npm project, generated tests... ran them. I was quite surprised.

Your version is the opposite of mine, in a very good way. Your one is mostly tests! My version's tests are... well, you'll see ;)

Both are surprisingly short. 700-ish for mine, ~1200-ish for yours (as far as the actual interpreter goes), right?

That seems like a lot of bang per buck for something as powerful as a Prolog interpreter! I don't know very much about Prolog though, so maybe there's a lot of crucial parts missing here.

At any rate the original is a teaching device, and the book[1] goes into some length on the limitations of Prolog, both this version and in general.

[0] Original source in Python: https://github.com/dhconnelly/paip-python

[1] Original original source in Lisp: https://norvig.github.io/paip-lisp/#/chapter11

> I don't know very much about Prolog though, so maybe there's a lot of crucial parts missing here.

An easy way to find out if your implementation works is to try out the `likes.pl`[1] example found in the SWI-Prolog "Getting started quickly"[0] documentation.

0 - https://www.swi-prolog.org/pldoc/man?section=quickstart

1 - https://raw.githubusercontent.com/SWI-Prolog/swipl-devel/mas...

If the project is about testing Claude's capabilities, the prompts will be more interesting to have.

It's not like it's a single shot generation or anything.

>> If the project is about testing Claude's capabilities, the prompts will be more interesting to have.

> It's not like it's a single shot generation or anything.

Since you've submitted a "Show HN: Vibe Prolog" and shared in the repo:

  I was working on something else and as a side effect 
  accidently vibe coded a prolog interpretor on my phone over 
  the weekend.

The GP's observation of the prompts being highly relevant is substantiated. Arguably even more so than the Python code committed.

Especially since there have been 82 commits in the span of 3 days.

EDIT:

And I am really interested in how you "vibe coded" the commit[0] which made this singular change:

   - uv run pytest tests/ttest_additional_builtins.py -v
   + uv run pytest tests/test_additional_builtins.py -v

0 - https://github.com/nlothian/Vibe-Prolog/commit/0bf4beba70dea...

Yeah I think I modified the Github action though the web interface for that or something? I don't recall.

I don't think this is a violation of my (self-imposed) rules - it's supporting config, not code.

> I don't think this is a violation of my (self-imposed) rules - it's supporting config, not code.

Before this[0] HN comment, I had seen no evidence of prompts used to affect change, only the commits pushed to GitHub. Maybe others had access to those prompts or I just missed them, either way I applaud you for having shown a sample of those used.

As to any determination of "violations", that is not my place. All I was hoping for was some evidence of "vibe coding" being employed. Without evidence, it is nigh on impossible for anyone to have confidence claims made are plausible outside of those making them.

0 - https://news.ycombinator.com/item?id=46044782

Maybe don't say things like this then?

  > So there are two reasonable explanations:

  >  A) There is no way to show prompts used when "vibe coding".
  >  B) This claim of "vibe coding" a Prolog is disingenuous.

The actual explanation is that it never occurred to me that this would interest anyone!

There's some more prompts here: https://github.com/nlothian/Vibe-Prolog/blob/main/docs/WORKF...

I do not think that [0] was vibe coded. Vibe coding something does not mean that everything is vibe coded, or at least I would have never expected it to be the case.

> I do not think that [0] was vibe coded. Vibe coding something does not mean that everything is vibe coded ...

The author claims in the project README.md[0]:

  I was working on something else and as a side effect 
  accidently vibe coded a prolog interpretor on my phone over 
  the weekend.

And has yet to show any evidence of "vibe coding", to the point where I chose one of many commits challenging this position.

So there are two reasonable explanations:

  A) There is no way to show prompts used when "vibe coding".
  B) This claim of "vibe coding" a Prolog is disingenuous.

Which do you think is the case?

0 - https://github.com/nlothian/Vibe-Prolog

README.md is definitely an output of an LLM.

As for the rest of the code, it could be. Are there any obvious patterns you have found to be LLM generated?

In any case, my point is that a typo probably does not have to use LLM to fix that single typo.

So you can use LLMs but it is probably not used for everything, for example this typo.

Have you considered asking for a version in assembly?

Not really. Do you mean targetting assembly output or written in assembly?