User asks if the PR review tool will not only review but also address identified issues and merge the PR to the main branch, seeing this as a way to reduce the PR bottleneck from AI-generated PRs.
Claude Code PR reviews are here, just $15–$25 a pop. I’m finding it hard to believe it’s worth it. Here’s what they said: "We've been running Code Review internally for months: on large PRs (over 1,000 lines changed), 84% get findings, averaging 7.5 issues. On small PRs under 50 lines, that drops to 31%, averaging 0.5 issues. Engineers largely agree with what it surfaces: less than 1% of findings are marked incorrect.” And yes, the stats sound good. It’s finding issues with very low false positive rates. But here's my concern, Anthropic has been dogfooding this internally for months. Meanwhile there are 3493 open Github issues classified as bugs against Claude Code today. I‘m finding bugs myself. I just can’t get Claude Code’s remote control feature to work for more than a few minutes at a time and from discussing it with others, I’m not alone. If their own tool can't keep their own product stable, what exactly are you paying $15–$25 a review for? I’m not convinced this is ready for teams shipping 10+ PRs a day. What do you think?