Why do LLMs like this kind of writing? Why did you need to “plainly” name the competition as opposed to “flamboyantly” naming them?
This is a cool idea but the README is absurdly not-to-the-point.
“I should not pretend other implementations don’t exist. I should name them plainly.” Unable to tell itself that it puts this in output so that it can tell itself there.
Same with the “not X but Y” style. It is self-steering introspection leaking into output for one reason or the other. Recent poor language use by model is probably an attempt to minimize this by being concise. But it really needs more thinking to solve problem and so it does some of it in output.
With GPT-5.6 I get reasonable results with "use natural, plain English". I tried "use ASD-STE100 Simplified Technical English" but I've found "natural plain English" to work better. Still not on the level I would write, but better.
In the mean time, I hear that Claude resists changing its writing style. A PO used Claude to write release notes and it was full of Claudeisms and tried changing the text into a TED talk. Then he prompted Claude to rewrite in ASD-STE100 Simplified Technical English, and Claude barely changed its writing. Then he switched to GPT-5.6 Terra, and GPT revised everything and made the text much better.
I'm willing to use (semi-)vibecoded projects under certain circumstances, but I like to know up front what I'm getting. The github contributors list makes that easy to see as long as people don't try to hide it.
Seems like you put a lot of thought and effort into it which is awesome, so the LLM readme was ruining first impressions. It's a shame but given the explosion in low effort projects it's important to stand out I think.