Claude Code 2.1.304 adds auto-mode setup, usage warnings and Code Review levels

Claude Code v2.1.304, released October 19, 2026, per its release notes:

• New: /auto-mode-setup, a guided setup that reads the project and recent sessions to teach auto mode about the environment, now also on Bedrock, Vertex, Foundry and gateway sessions; /claude-api managed-agents-secure, which compares a Managed Agents agent's settings with Anthropic's security guide; and a warning on Pro and Max plans when resuming an inactive conversation would use a noticeable share of the 5-hour usage limit.
• Code Review: admins can choose Quick, Standard (default) or Deep review levels per organization or repository, and can review every repository in a GitHub organization, including future ones. Pull requests opened right after Code Review is switched on are no longer skipped.
• Mods: new hooks and fields for mods (decider: "person", e.isReadOnly), and settings.read hooks now receive each settings layer.
• Fixes include: claude --worktree no longer removes a worktree and branch without asking when HEAD is detached or on another branch, which could lose commits; rm -r of a mounted drive now gets the safety prompt; sandboxed commands failing on Linux with bracketed deny-read globs; and several Remote Control, Cowork and subagent issues.
• Tool search is now on by default behind a custom ANTHROPIC_BASE_URL, and the "(research preview)" label was removed from the /fast dialog on Opus 5.5.