Anthropic announced that auto mode will become the default setting for its Claude Code platform starting August 14, affecting Pro, Max and Team accounts. Developers will no longer need to manually approve each action AI takes during programming tasks, as Claude Code will automatically process unless an action is deemed ‘irreversible, destructive, or aimed outside your environment’. The company first introduced auto mode in March as a test feature. Testing showed auto mode caught 89% of harmful actions compared to just 13.6% detected by human reviewers. Users tend to approve nearly all permission prompts manually, making automated oversight more effective. The feature works by first checking for a connector for jobs like Slack or Google Calendar; if none exists, it falls back to controlling your screen, opening files, navigating browsers, typing into forms, and clicking menus. It always asks permission before touching a new application. The feature pairs with Dispatch, Anthropic's task-assignment tool, allowing users to delegate work from their phone and set recurring tasks. Auto mode is available for Pro and Max subscribers, requiring the Claude desktop app installed on macOS. Claude Code Head Boris Cherny shared on X that his team has used auto mode exclusively for months, saying he “couldn’t imagine going back to permission prompts.”