New proposal for LLM-on-LLM wars: Have an automated system that retrieves every front page URL link on HN and feeds it into a "smart" LLM for analysis of whether or not the site html, css, JS and general template layout match that of a known LLM generation engine. Then give it a "probably vibe coded slop" rank number. If you can't beat 'em, let them fight...

▲

zrail an hour ago | parent [-]

I haven't tried it personally but my understanding from friends is that Pangram is very accurate, at least for now.

https://www.pangram.com/

▲

_joel an hour ago | parent | next [-]

Ironic that site looks like it's AI generated too

	▲	koito17 31 minutes ago \| parent [-]
		It looks close to the kind of documentation artifacts GPT 5.6 Sol would generate for me. That style is still rare enough that I prefer it over OP's "we installed shadcn + Tailwind and look like every seed-stage startup SaaS from 2024" style.

▲

walrus01 an hour ago | parent | prev [-]

As a paid service it probably works great. In the spirit of weaponizing vibecoded slop vs vibecoded slop, I'm fairly sure that I could 'teach' locally hosted Qwen 3.8-Flash-Next to identify vibe coded slop, with a thorough enough system prompt and access to like, playwright and headless chromium.

Or since a lot of vibe coded static sites are just plain html, cs, js, a first try attempt via wget to mirror the first 500KB of the site and then analyze it.