LLM's has their advantages on code scrutiny, especially on security issues. They can also suggest easyly on some code evolutions (eg: detection/replavement of some obsolted functions)
However, that is very easy to fall rapidly in "anything bloated" if you "prompt" a while on that kind of sh*t. /I falled in it while my first attempt was to correct bugues on an old code. Then reverted all changes, and deleted a brand new code entierely due to AI on the fu**in GitHub where it given capabilities to a web "framework" where AI has certainly a large responsability in coding. Be Careful! /However, I'm part of a "musicals" company of mentally impaired persons, and we currently writing a new spectacle "on/about AI", but especially *without any help from* / as we litterally destroys its reputation. Issued most probaly 2028... Cheers, Thomas Harding -- http://tharding.fr "Opération Que Du Plaisir" en PDF. sous licence CC-By-ND 4.0 Une œuvre surréaliste féministe révolutionnaire sur La Pataguerre : une opération militaire pour filles... Le 10 septembre 2026 19:35:28 GMT+02:00, Ineiev <[email protected]> a écrit : >On Thu, Sep 10, 2026 at 06:20:02AM +0000, निरंजन wrote: >> >> I am >> posting this here as I want to know how Savannah would treat >> vibe-code submissions. > >The FSF is working on the rules about accepting code generated >by LLMs; until they are ready, GNU maintainers are asked to avoid >installing any LLM-generated code in GNU packages. > >It would be nice if you waited for the FSF as well. > >> If acceptable, should they go to Savannah GNU or Savannah >> non-GNU? > >If your contributions went to GNU Savannah, you wouldn't >ask this question; therefore, savannah.nongnu.org. >
