Claude Code launched/ultrareview, multi-agent cloud parallel code review to find bugs

ME News: On April 23 (UTC+8), according to Dongcha Beating’s monitoring, Anthropic launched /ultrareview\ (research preview) in Claude Code, a cloud-based multi-agent code review feature. Users enter /ultrareview\ in the CLI, and the system starts a set of review agents in a remote sandbox to check the diff between the current branch and the default branch in parallel (including uncommitted changes). Users can also pass a PR number to review a GitHub PR directly. The entire process does not consume local resources, takes about 5 to 10 minutes, and the results are returned to the session as notifications.

The core difference from the local /review is the verification mechanism: each finding is reproduced and confirmed by an independent agent, focusing on real bugs rather than code style suggestions. The official documentation describes the two as tools for different stages: /review\ is for quick feedback during the coding process, while /ultrareview\ is for in-depth review of critical changes (e.g., authentication, data migration) before merging.

In terms of billing, /ultrareview\ is charged under extra usage and does not consume plan usage. Pro and Max users each receive 3 free quotas (one-time, non-refreshing) before May 5; after that, each use costs approximately $5 to $20 depending on the scope of the changes. Team and Enterprise users have no free quotas.

The feature requires authentication via a Claude.ai account and cannot be used on Amazon Bedrock, Google Cloud Vertex AI, Microsoft Foundry, or organizations with zero data retention enabled. (Source: BlockBeats)