I've been using codex and never had a compile time error by the time it finishes. Maybe add to your agents to run TS compiler, lint and format before he finish and only stop when all passes.
That’s a separate tool though. You don’t want to have to open another terminal to git diff every 30 seconds and then give feedback. Much better UX when it’s inline.
My main hooks are desktop notifications when Claude requires input or finishes a task. So I can go do other things while it churns and know immediately when it needs me.
You could easily guard against bullshit issues. So you can focus on what matters. If the issue is legit goes ahead to a human reviewer. If is run of the mill ai low quality or irrelevant issue, just close. Or even nicer: let the person that opened the issue to "argue" with the ai to further explain that is legit issue for false positives.
reply