I tested the new OpenAI Codex features on a real Python codebase, and it’s the strongest Claude Code rival yet

Late last month, OpenAI shipped a new product they dubbed “Codex for (almost) everything.” Its goal was to shift Codex from a code editing product to a general-purpose tool. The “Codex for (almost) everything” launch included features such as computer use, an in-app browser, PR review, SSH connections to remote dev boxes, and over 90 new plugins. More than 3 million developers use Codex every week, and many of them probably opened the app the next morning and didn’t notice anything different. I noticed. And now I have opinions.