Claude Code 的自动模式

A new permissions mode called auto mode has been introduced in Claude Code as an alternative to <code>--dangerously-skip-permissions</code>. In auto mode, Claude makes permission decisions on behalf of the user, with safeguards monitoring actions before they are executed. These safeguards appear to be implemented using Claude Sonnet 4.6, as described in the documentation. Before each action runs, a separate classifier model reviews the conversation to determine if the action aligns with the user's request. This new feature aims to enhance user experience by reducing unnecessary permission prompts while ensuring the safety of actions.