Everything I could decide myself is decided. What's left needs
you. Each one has a recommendation from Fable — if you agree, pick it and
press Done. "I'm not sure" is a real answer and becomes work for me.
You don't need to open the session — pressing Done sends your answer back and it picks up.
1Our main copy of the code can't take updates right now. 21 files are sitting half-merged from an interrupted operation about two hours ago, and git refuses to pull until someone decides how those files should end up. Nothing is lost and nothing is on fire — everything is backed up, and the individual worktrees you and the agents actually work in are unaffected because they update themselves directly. But the shared copy stays 109 commits behind until this is cleared. How do you want it handled?
Fable recommends: You (or a session you point at it) resolve the 21 files, then I fast-forward and confirm The interrupted operation looks like our own overnight maintenance job restoring stashed edits, so those 21 files may hold real in-progress work — and 4 of them differ from the latest code, so option 2 would silently discard something. The maintenance code's own rule says this exact situation must stop and ask a person rather than self-heal. There's no urgency: your worktrees update themselves independently, so nothing you're working on is blocked. Tried first, unsuccessfully: receipt f0df8a1e3 — asked twice with options reversed, agreed both times (high/high). It found the hygiene module's own doctrine at git_hygiene.py:52-57: 'unmerged files' is deliberately NOT one of the three failures that license an automatic heal, so this state is designed to route to a human. It also corrected me that worktrees branch off origin/main which fetch still updates, so the blast radius is much smaller than it first looked.