BREAKING
Claude Code Adds Review Effort Levels
Five Levels From Low to Ultra
1Low: single pass
2Medium: in-context
3High: fresh agents
4X-high: ripple check
5Ultra: agent fleet
0%
PRs before
0%
PRs after
0%
large PRs flagged
0$
max per PR
0min
typical run
0x
comments large PR
When to Use Which Mode
Lowroutine
Fast and cheap
Everyday pushes
Ultrahigh-risk
Independent verification
Auth or billing changes
Research Preview on Team and Enterprise
AI NEWS BLITZ
Anthropic now lets developers dial the depth of Claude Code's code reviews.