Skip to content

Claude Code's /code-review drops cleanup nits by default

Version 2.1.328 moves simplification and efficiency suggestions out of the default review, into a separate /simplify command. The same release expands auto mode to third-party clouds.

Source: Anthropic

Anthropic shipped Claude Code 2.1.328 on 19 November, changing what /code-review reports by default. On most models it now flags only bugs and CLAUDE.md violations; reuse, simplification and efficiency suggestions move to a separate /simplify command. Each finding is now labelled Important, Pre-existing or Nit, and a clean run ends with “Code review found no issues.” instead of the old summary text.

The same release extends auto mode, the setting that controls whether interactive sessions start without prompting for permissions, to Claude Opus 4.6 and Sonnet 4.6 running on Amazon Bedrock, Google Cloud’s Agent Platform and Microsoft Foundry. Plan mode and Bash prompts on those third-party providers now offer auto mode the way newer models already do.

disableAutoMode: true

Why it matters: anyone with a script, hook or habit built around /code-review catching style and efficiency issues needs to start calling /simplify separately to get that feedback back. Anyone running Claude Code against Bedrock, Google Cloud or Microsoft Foundry should expect a new default around permission prompts, not just on Anthropic’s own API.

The caveat: this is one change inside a release with 91 other fixes, easy to miss if you only skim the headline items.

Share

Related reading

More from The Wire

All briefs

A Linux KVM bug let one VM read 127GB of host memory

A bug bounty researcher used a four-year-old stale pointer in nested virtualization to read host kernel memory from inside an ordinary cloud VM. DigitalOcean disclosed and patched it as CVE-2026-99312.

via DigitalOcean