PSA: Claude Code enabling auto mode as default next week, Anthropic says

Anthropic said Friday that auto mode will become the default permission setting for Claude Code users starting August 14. The change applies to Pro, Max, and Team users unless a user or administrator has pinned another setting. Auto mode replaces repeated approval prompts with a classifier that checks each tool call for irreversible, destructive, or out-of-bounds actions. If the classifier blocks something, Claude can try a safer route or ask permission. After repeated blocks, the session falls back to manual approval. Anthropic said it will no longer charge for the small number of extra tokens per tool call used for the auto mode classifier. Anthropic said its study of 1,053 paid testers found people caught 13.6% of dangerous commands compared with 89% for auto mode. Human performance fell to about 5% after 50 prompts. Anthropic cautioned that classifiers cannot eliminate risk and recommended human review for production changes. The company said auto mode users at Team and Enterprise customers ship about 25% more pull requests.