What does /auto-mode-setup do in Claude Code 2.1.304?

The /auto-mode-setup command in Claude Code 2.1.304 is a guided setup that reads your project and your recent sessions, then teaches auto mode about your environment. According to the release notes, it is now also available on Amazon Bedrock, Google Vertex, Microsoft Foundry, the Claude apps gateway, and sessions with telemetry turned off.

Which review levels does Claude Code Review offer?

Claude Code 2.1.304 adds three review levels to Code Review: Quick, Standard and Deep, with Standard as the default. Organization owners choose the level for the whole organization and can give individual repositories their own level. A new admin option can also review every repository in a GitHub organization, including repositories added later.

Why does Claude Code warn me before resuming an old conversation?

Claude Code 2.1.304 shows a warning on Pro and Max plans when resuming an inactive conversation would use a noticeable share of the 5-hour usage limit. Users can resume anyway, or start a new conversation; the release notes say Claude can look up the previous conversation if you refer to it.

Did Claude Code 2.1.304 fix any data-loss bugs?

Yes. Claude Code 2.1.304 fixes claude --worktree removing the worktree and its branch without asking at exit when HEAD was detached or on another branch, which could lose commits. It also fixes CLAUDE_CODE_TRANSCRIPT_LOCAL_GC dropping messages from the saved transcript when the disk filled up during trimming.

Anthropic · 2026-10-19 · major

Claude Code 2.1.304 — guided auto mode setup and Code Review levels

Claude Code 2.1.304 adds /auto-mode-setup, a guided setup that teaches auto mode about your project. Code Review gets Quick, Standard and Deep levels, and Pro and Max users are warned before a costly resume.

Claude Code 2.1.304 makes auto mode easier to set up and lets teams choose how deep Code Review goes.

Quick facts

Version	v2.1.304
Maker	Anthropic
Released	2026-10-19
New command	/auto-mode-setup (also on Bedrock, Vertex, Foundry)
New skill command	/claude-api managed-agents-secure
Code Review levels	Quick, Standard (default), Deep

What is it?

A guided /auto-mode-setup command arrives in Claude Code 2.1.304. It reads your project and recent sessions to teach auto mode about your environment, and it now also works on Bedrock, Vertex, Foundry, the Claude apps gateway and telemetry-disabled sessions. A new /claude-api managed-agents-secure command compares a Managed Agents agent's settings with Anthropic's published security guide without changing anything.

How does it work?

Mods get finer permission control: a tool.check answer can set decider: "person", so only the permission dialog or a host callback can approve the request, and the event now carries e.isReadOnly for read-only calls. On the Code Review side, owners pick a review level for the whole organization or per repository, and can turn on review for every repository in a GitHub organization, including ones added later.

Why does it matter?

Pro and Max users now see a warning when resuming an inactive conversation would use a noticeable share of the 5-hour usage limit, with the option to start fresh instead. The release also fixes claude --worktree deleting a worktree and its branch without asking when HEAD was detached, which could lose commits, and adds the rm -r safety prompt for mounted drives such as /Volumes and WSL's /mnt/c.

Who is it for?

Claude Code users, teams using auto mode or Code Review, mod authors

Frequently asked questions

What does /auto-mode-setup do in Claude Code 2.1.304?: The /auto-mode-setup command in Claude Code 2.1.304 is a guided setup that reads your project and your recent sessions, then teaches auto mode about your environment. According to the release notes, it is now also available on Amazon Bedrock, Google Vertex, Microsoft Foundry, the Claude apps gateway, and sessions with telemetry turned off.
Which review levels does Claude Code Review offer?: Claude Code 2.1.304 adds three review levels to Code Review: Quick, Standard and Deep, with Standard as the default. Organization owners choose the level for the whole organization and can give individual repositories their own level. A new admin option can also review every repository in a GitHub organization, including repositories added later.
Why does Claude Code warn me before resuming an old conversation?: Claude Code 2.1.304 shows a warning on Pro and Max plans when resuming an inactive conversation would use a noticeable share of the 5-hour usage limit. Users can resume anyway, or start a new conversation; the release notes say Claude can look up the previous conversation if you refer to it.
Did Claude Code 2.1.304 fix any data-loss bugs?: Yes. Claude Code 2.1.304 fixes claude --worktree removing the worktree and its branch without asking at exit when HEAD was detached or on another branch, which could lose commits. It also fixes CLAUDE_CODE_TRANSCRIPT_LOCAL_GC dropping messages from the saved transcript when the disk filled up during trimming.

Try it

npm i -g @anthropic-ai/claude-code@2.1.304