Why not just have the LLM write the PR in the first place? Because LLMs are imperfect tools. At least for the foreseeable future the human in the loop is still important
I am assuming that, for the vast majority of code changes moving forward, the PR will be written by an LLM in the first place.
I am assuming that, for the vast majority of code changes moving forward, the PR will be written by an LLM in the first place.