Erik-JanMar 17

We're starting to see increasing maintainer burden in one of our (smallish) open source projects due to GenAI / LLM code and comment "contributions" (or rather: encumbrances)

As a result we want to ban those altogether. Any tips on how, specifically, to do this? Is there some metadata we can put on our repo? Do we just put it in the readme and in the contributing guidelines? Do we add an issue template / PR template? All of the above?

#LLM #AI #Slop #OpenSource #GitHub

Show threadJacob Scott
@erikjan vouch? https://github.com/mitchellh/vouch
GitHub - mitchellh/vouch: A community trust management system based on explicit vouches to participate.

A community trust management system based on explicit vouches to participate. - mitchellh/vouch

GitHub
Mar 17 at 12:29pmWeb
Show threadErik-JanMar 17
@_wurli ah yeah, that looks interesting! Perhaps a bit too involved for the level of the project I'm talking about (±50 GitHub stars, only a few external contributions)?
Show threadJacob ScottMar 17
@erikjan that’s fair! Maybe a simple CONTRIBUTING.md would be enough then, if your contributors are willing to follow it
Show threadErik-JanMar 17
@_wurli yeah I guess this will be the solution and then we can just reject any LLM generated stuff. The annoying thing is then we have to "accuse" accounts of using LLMs which is not 100% foolproof either
Show thread⊥ᵒᵚ Cᵸᵎᶺᵋᶫ∸ᵒᵘ ☑️Mar 17
@erikjan @_wurli get off #github, whose owner #microaoft is pushing the AI slop and so has no interest in helping?
#codeburg seems to have good defenses, still fully #git compatible