Post

ok I just did this on a python print("Hello, World!") and the code-review max was like "are you kidding, no changes are needed" so idk what's going on here

there's a lot of skill issues out there, I think

If you input premade code and ask for a check you will get a different outcome than if you ask it to write the code. Since most people use Claude to write code the problem becomes obvious. So, yes, there appears to be a skill issue, but not with the prompter in the reddit post.

This isn't the first person to have this issue, so I'm not sure what y'all are proving here. It's not accurate enough, often produces bloated product, and will prompt you into engagement where it can bc it's programmed to. What expert disputes this?