How do you handle multi-session tasks with AI coding assistants?

Curious how others deal with this: you start a refactoring task with your AI coding assistant, close the terminal, come back – and it has no memory of what you were doing.

Codex v0.128.0 just shipped goal which persists objectives across sessions. You goal create an objective, and it survives terminal restarts. goal resume restores execution context, not just the goal text.

Also added pet – an animated companion that shows agent background state (active, stuck, idle). Sounds silly but it is basically agent observability without tailing logs.

  • Are you using any workarounds for session persistence in other tools (Claude Code, Cursor)?

  • How do you monitor long-running agent tasks today?

  • Does “emotional design” in dev tools matter to you, or is it noise?