Claude Code Auto Mode Now Default for Paid Plans

Claude Code is about to stop asking permission for most things. Starting August 14, Anthropic will switch on Auto Mode by default for anyone on a Pro, Max, or Team plan, letting the coding tool run through tasks without pausing for manual sign-off at each step. Only Enterprise customers will still need to opt in manually. A classifier built into the system decides whether a given action is risky or hard to undo, and only flags those cases for human review, everything else just runs.

The shift comes after Anthropic tested the feature with 1,053 paid users alongside its own red-teaming exercises. The company says Auto Mode matched or beat manual approval workflows on safety while letting teams ship roughly 25 percent more pull requests. Anthropic is also pitching the mode as a defense against prompt injection attacks, where hidden instructions buried in code try to redirect the agent away from what the user actually asked for. That claim comes with numbers attached, including results from an outside audit and at least one internal incident where the system caught something before it became a problem.