Analysis
This weekly update for Claude Code brings some of the most exciting quality-of-life and performance improvements for developers! The immediate standout is the introduction of the "/ultrareview" command, which utilizes parallel agents in the cloud to provide comprehensive, multi-perspective code reviews. Combined with the powerful Opus 4.7 model becoming the default and a highly requested interactive inference slider, this update represents a massive leap forward in streamlined AI-assisted coding.
Key Takeaways
- •Claude Opus 4.7 is now the default model, and a new 'xhigh' reasoning level has been introduced between 'high' and 'max'.
- •The new interactive '/effort' command allows developers to dynamically adjust inference power on the fly using arrow keys during a session.
- •Auto mode is now permanently unlocked for Max subscribers without the need for any startup flags.
Reference / Citation
View Original"/ultrareview is a new slash command for code review. Rather than calling a single local review agent, it runs multiple agents in parallel on Anthropic's cloud to comprehensively check the code all at once."