
Okay, but Claude Code is an agent harness and behind the scenes when it's "just fixing that bug" it's using loops. It does the work then checks its work... etc.
If you aren't using Claude Code's finer mechanisms (creating skills and such for repeatable actions) you're only using Claude Code for its most basic surface tool.
If that's how you want to work, that's fine, but this isn't exactly a flex.
English




























