What happens next is the pattern that shows up on almost every team that adopted agents seriously. Review turns into skim. The reviewer reads the PR description (written by the agent), checks that CI is green (on tests written by the agent, against a spec the agent summarized from the ticket it was handed), scrolls past the longest file because it's mostly generated boilerplate and approves. It looks like review. It's a signature.
I'm not saying that to shame anyone, it's rational. Reading every generated line is not a strategy that scales and pretending it is just means the skim happens in secret. The honest version is to admit the line-by-line read is gone for most changes and ask what replaced it.
Top comments (2)
The allow/ask/deny split is a useful framing. I’d add one reviewer stop condition: if a PR changes files outside the ticket, requests new tool permissions, or lacks a regression test with a failing example, route it to human review rather than auto-merge. That keeps the agent useful without making a confident summary the approval signal.
还是Threerouter desktop soft 好用