Auto Mode by Claude Code - Let Claude make permission decisions on your behalf

Claude Code’s new auto mode lets Claude approve file writes and bash commands for you. A classifier checks each action: safe ones run automatically, risky ones get blocked and handled differently. Use in isolated environments.

Add a comment

Replies

Best

been using claude heavily for building and the jump to computer use is interesting how does it handles ambiguous instructions where the right next click isnt obvious.

I am still trying to compare between the latest auto mode by claude and openclaw. Any advice or learnings so far?

Been using Claude heavily for solo development — entire products built in days with Claude Code. What separates it from other models is how it handles context over long sessions and pushes back when something doesn't make sense, rather than just agreeing and generating garbage. For indie makers building without a team, that behavior matters more than raw speed. The auto mode launch is the direction I wanted — less friction in the loop.

Using a lot Claude at the moment. It really helps me developing faster and cleaner.

I’ve been using —dangerously-skip-permissions ever since Opus 4.6 rolled out and never looked back. The model is so good permissions aren’t even really needed anymore more. You’re wasting so much time if you’re approving everything.