AI อะไรเนี่ย

Tools

Claude Code Introduces Auto Mode for Automated Permissions

Claude Code Introduces Auto Mode for Automated Permissions

Today, Claude Code is rolling out a significant enhancement for developers: Auto mode. This new permissions mode empowers Claude to make permission decisions automatically, streamlining workflows while still maintaining crucial safeguards. It's designed to reduce interruptions during larger tasks, offering a more efficient yet secure alternative to constant manual approvals or the risky --dangerously-skip-permissions option.

What is Auto Mode?

At its core, Auto mode for Claude Code allows Claude to automatically approve safe actions, ensuring your AI assistant can progress without needing constant human intervention. Traditionally, Claude Code's default permissions are conservative, requiring approval for every file write or bash command. While safe, this can hinder productivity for extensive projects.

Auto mode introduces a "middle path." Before each tool call runs, a sophisticated classifier reviews it to identify potentially destructive actions, such as mass file deletion, sensitive data exfiltration, or malicious code execution. Safe actions proceed automatically, while risky ones are blocked, prompting Claude to find a different, safer approach. Should Claude persistently attempt blocked actions, a user permission prompt will eventually be triggered, putting you back in control. This innovative feature is compatible with both Claude Sonnet 4.6 and Opus 4.6 models. You can learn more about this exciting development in the Auto mode announcement.

Why It Matters for Developers

For developers, Auto mode translates directly to increased efficiency and a smoother workflow. Imagine initiating a complex coding task and knowing Claude can handle a significant portion of it autonomously, without requiring you to constantly approve individual steps. This is particularly beneficial for long-running processes where frequent interruptions can derail focus.

While it reduces risk compared to completely skipping permissions, it's important to remember that Auto mode is currently a research preview. It aims to strike a balance between speed and security, enabling more hands-off development while mitigating severe risks. This makes Claude Code an even more powerful assistant for tackling ambitious projects.

Getting Started with Auto Mode

Auto mode is currently available as a research preview for users on the Claude Team plan and will be rolling out to Enterprise plan and API users in the coming days.

For developers, enabling Auto mode is straightforward:

  • CLI: Run claude --enable-auto-mode, then cycle to it using Shift+Tab.
  • Desktop App / VS Code Extension: Toggle Auto mode on in Settings -> Claude Code, then select it from the permission mode drop-down within a session.

For administrators, Auto mode can be disabled for the CLI and VS Code extension by setting "disableAutoMode": "disable" in your managed settings. On the Claude desktop app, it is disabled by default but can be toggled on via Organization Settings. While Auto mode significantly reduces risk, it doesn't eliminate it entirely; using it in isolated environments is still recommended. Do note that Auto mode may have a small impact on token consumption, cost, and latency for tool calls.

Read more: Explore Claude Code's capabilities and revolutionize your development workflow.