I manage 3-4 real projects these days, and paying constant attention to all of them is exhausting. Even with agents doing the work, I was still the one resolving every decision from Claude or Codex stopped on — including the trivial ones. That's the part I wanted to fix pmtui runs an auto mode on top of the session. When a session stops on a decision, a 2nd claude/codex reads it, and if it's routine it answers — with the goal I set as context. You can think of it as using Claude to decide for another Claude when the decision doesn't need a human. The other half is keeping sessions alive without me. A daemon watches each tmux session and nudges it on a cadence I choose, building each prompt from that agent's own last status and my goal, instead of a generic "continue" that makes it re-plan or quit early.

pmtui

Autopilot for long-running AI sessions in terminal

Autopilot for long-running AI sessions in terminal