Overview
- Anthropic will switch Claude Code’s Auto mode to the default for Pro, Max and Team accounts starting August 14, and existing users will get a one‑time prompt while admins can pin or disable the setting.
- Auto mode routes every tool call through a safety classifier that blocks irreversible, destructive, or out‑of‑environment actions and falls back to manual approval after repeated blocks.
- In a controlled study of 1,053 paid testers Anthropic reported the classifier blocked about 89% of dangerous commands versus 13.6% caught by human reviewers, and independent tests found Auto mode resisted prompt‑injection attacks.
- Anthropic says Auto mode boosts productivity—Teams and Enterprise users on the feature shipped about 25% more pull requests—and named customers including Adobe, Nuro, Gusto and Garner Health already use it in production.
- Auto mode will remain opt‑in for Claude Enterprise, the Claude API and major cloud agent platforms for now, and Anthropic will not charge Pro, Max and Team users for the extra classifier tokens while urging manual review for high‑stakes production changes.