"Dispatching agents on a pull request, Code Review dispatches a team of agents that look for bugs in parallel, verify bugs to filter out false positives, and rank bugs by severity, according to Anthropic. The result appears in the pull request as a single, high-signal overview comment, plus in-line comments for specific bugs."
"On large pull requests (more than 1,000 lines changed), 84% get findings, averaging 7.5 issues. On small pull requests of fewer than 50 lines, the rate of findings drops to 31%, averaging 0.5 issues. Anthropic has found that its engineers mostly agree with what Code Review surfaces, marking less than 1% of findings as incorrect."
Anthropic introduced Code Review, a new feature for Claude Code available in research preview for Claude for Teams and Enterprise customers. The system deploys multiple agents to analyze pull requests in parallel, identifying bugs, verifying findings to eliminate false positives, and ranking issues by severity. Results appear as a consolidated overview comment plus inline annotations for specific bugs. Average review time is approximately 20 minutes. Internal testing shows the system detects issues in 84% of large pull requests (over 1,000 lines changed), averaging 7.5 issues per review. Smaller pull requests under 50 lines show 31% detection rates with 0.5 average issues. Anthropic engineers validate findings with less than 1% disagreement rate.
#code-review-automation #multi-agent-ai-systems #bug-detection #software-development-tools #claude-ai
Read at InfoWorld
Unable to calculate read time
Collection
[
|
...
]