Anthropic · 2026-10-15 · notable
How the Claude Code team does code review — seven layers, few humans
Anthropic's Claude Code team explains its seven-layer review process for code that Claude writes about 90% of. Merged PRs more than doubled from June to September while the revert share fell by more than 40%.
Anthropic shows how it reviews code when Claude writes most of it: automated layers first, humans for the risky calls.
What is it?
"How the Claude Code team does code review" is a 15 October 2026 engineering post by Addy Osmani, Morgan Westlee Lunt and Qing Wang. Claude writes about 90% of Claude Code's own code, so the team rebuilt review around one goal: less human review time with no rise in reverts or incidents.
How does it work?
Review now runs in seven layers: Claude verifies its own work and a fresh-context subagent reviews the diff before a PR; parallel agents hunt bugs and a second group verifies each finding; a blocking security review; required CI; risk-based human approval set by a classifier; a merge queue with flags and staged releases; and fast rollback. Past incidents serve as an eval set, and reviewer changes that lower the score are dropped.
Why does it matter?
The team reports that one-round passes rose from 65% to 78% and PRs needing five or more rounds fell from 8.9% to 1.5%. Low-risk changes such as docs merge with no human approval once checks pass, while billing, the auto-updater and CI workflows always need a person. The post also says Code Review for Claude Code is now generally available on Team and Enterprise plans.
Who is it for?
engineering leads adopting coding agents
Try it
Run /code-review in a Claude Code session