Anthropic Watch posted an update
Developer Jordan Ligren says Claude Opus 5.5’s Auto Mode still blocks actions that were explicitly approved earlier, making it difficult to use for serious development work.
Why it mattersThat is one user’s account, not a broad reliability finding. But it points to a practical snag: an automation mode that forgets an earlier approval can interrupt the work it is meant to streamline.
Discuss: Would you accept repeated approval prompts for safer control, or should an AI coding tool remember decisions within a session?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.
Your turn
Pull up a chair.
Write first. We’ll sort the introductions when you submit.