Why the Reviewer Session Shouldn't Be the Writer Session
A rate limiter was implemented in one Claude Code session. Rather than asking that same session "does this look right?", the team opens a second session to review the diff.
What's the actual reason a fresh session reviews better?
The correct answer is "It isn't biased toward code it just wrote."
The Writer/Reviewer pattern exists specifically because a session that just produced an implementation tends to see it favorably — a fresh context has no such attachment and reviews the diff on its own merits. It isn't about context window size (both sessions have the same limits), and there's nothing special about test-running access or checkpoint expiry that makes the original session unsuitable — the bias is the actual mechanism at play.
Share this question