It's one of those things you shouldn't trust LLMs to such an extent; that part should be very solid because the consequences of bad practices are getting to front page of hacker news :)

depends what LLM you are using but most frontier models have seen almost every github/doc/best practices its very hard to get something like supabase/lovable type of mess unless you purposely prompt it to be bad without much inner knowledge but even then it is rectifiable with the right prompts