Anthropic is making a change: Starting August 14, auto mode will become the default permission mode in Claude Code for Pro, Max, and Team users, the company announced via its bot.
Auto Mode automatically reviews shell commands and actions using a separate classifier. In testing, the system caught 89% of dangerous commands—manual user approvals only identified 14%.
What's changing
- Automatic security review becomes the default for paying users
- Classifier runs in the background without requiring manual approval for each command
- Significantly higher detection rate for dangerous operations versus manual review
The shift aims to protect users from unintended or malicious shell commands without slowing workflows. For German developers and companies using Claude Code in product development, this could mean a noticeable security improvement—assuming the classifier performs reliably in practice and doesn't generate excessive false positives.
Sources
Editorially owned by Ideal Syka. Sources and method: Newsroom & method. Tips and corrections: ai@i6eal.de.
