Use /rewind to Undo Bad Turns Without Wasting Tokens
When Claude heads down the wrong path, your instinct might be to explain what went wrong and ask it to try again. But that means Claude processes all the bad output plus your correction, burning tokens on context that shouldn't exist.
Instead, rewind:
> /rewind
This restores both the conversation and your code to a previous checkpoint. Claude never sees the bad turn, so you don't pay for it on subsequent messages.
You can also double-tap Escape to trigger the same thing. Press Escape once to stop Claude mid-response, then press it again to rewind to the last checkpoint.
This is particularly valuable during complex refactoring or multi-step implementations. If Claude makes a wrong architectural choice three steps in, rewinding gets you back to the decision point without carrying forward thousands of tokens of incorrect code and explanations.
For maximum efficiency, combine /rewind with plan mode. Press Shift+Tab to enter plan mode, let Claude propose an approach, and only proceed with implementation once you've approved the plan. If the plan is wrong, rewind is cheap. If the implementation goes sideways, you can rewind back to the approved plan.
Don't argue with a bad turn. Rewind it and start fresh.
Log in to leave a comment.
When typing feels slow — describing a complex bug, explaining architecture, or thinking through a problem out loud — press Option+V to switch to voice input. Speak naturally and Claude Code transcribes your words into a prompt, so you can describe what you need at the speed of thought.
When Claude is heading down the wrong path — editing the wrong file, writing code you don't want, or giving a long explanation you don't need — press Escape to stop it immediately. You keep everything it did up to that point and can redirect with a new prompt.
Closed a session and realized you weren't done? Pass --continue (or -c) when launching Claude Code to pick up exactly where you left off — same context, same files, same conversation history — without re-explaining what you were working on.