Anthropic will make Claude Code’s Auto Mode default for Pro, Max, and Team users on August 14, shifting coding oversight toward AI-managed approvals.

Anthropic says Claude Opus 5 and Auto Mode reduced browser prompt injection success to zero in internal tests, a notable claim for AI agents.

Anthropic introduced a new 'auto mode' for Claude Code that uses AI classifiers to autonomously approve safe actions and block risky ones, reducing developer approval fatigue while preventing incidents like accidental file deletion or credential exposure.