I must be using Codex wrong. I evaluated it with a task to do a pretty simple, mechanical string replacement across many files (moving from a prop spread in JSX to standard props, but only when the value being spread is a subscript of an object). It simply couldn't do it, and it wasn't even close. It was getting the syntax wrong, trying to fix it by deleting the code, then moving on to other files. Sonnet 4.1 wasn't perfect, but I was able to point out its errors and it fixed them and avoided doing it again.
I will say, Claude does seem to need a verbose prompt. Often I'll write my prompts as tasks in Notion and have it pull then via MCP (which is nice, because it tracks and documents its work in the process). But once you've given it a few paragraphs about the why and how, my experience is that it's pretty self sufficient. Granted, I'm using Cursor and not CC; I don't know if that makes much of a difference.
lkey
Codex cannot fail, it contains multitudes beyond your imagining. Nay, it can only be failed. Continue internalizing that the problem is you, not the tool. Perhaps a small infusion of one trillion USD would unlock it and your potential?
I will say, Claude does seem to need a verbose prompt. Often I'll write my prompts as tasks in Notion and have it pull then via MCP (which is nice, because it tracks and documents its work in the process). But once you've given it a few paragraphs about the why and how, my experience is that it's pretty self sufficient. Granted, I'm using Cursor and not CC; I don't know if that makes much of a difference.