How is an llm supposed to identify an llm-generated bullshit issue...? It's the fox guarding the henhouse.

Just try and you'll see if it can work. Just copy paste some of these issues give context of the project and ask if makes sense