Claude Code will make auto mode the default permission setting
Anthropic
Anthropic announced that Claude Code's auto mode, which routes tool calls through a safety classifier instead of requiring manual approval for each action, will become the default for Pro, Max, and Team users starting August 14. Anthropic cited a study of 1,053 testers in which the classifier blocked 89% of dangerous commands versus 13.6% caught by human reviewers, and said it will stop charging for the classifier's token overhead on these plans.
Why it matters
Signals Anthropic's confidence in automated safety classification over manual human approval for agentic coding actions, a significant default-behavior shift for a widely used developer tool.
Importance: 3/5
Default-behavior change for a widely used dev tool, backed by a large-scale safety comparison study.