Tmux Claude Hatch — Development skill for Claude Code

tmux-claude-hatch

[](https://youtu.be/NnTV6r4l5D0)

A tmux plugin (plus an optional Claude Code plugin) that runs a Claude Code session in a popup for each project directory. Open the hatch, hand Claude the work, close it and get back to your editor. Each session lives in its own nested tmux session, so closing the hatch never interrupts it. When you want to check in, an `fzf` picker lists every running session, shows what each one is doing, and jumps you straight to it. And when a session finishes or needs your attention, it rings a bell that highlights the window you launched it from — so you know where to go without even opening the picker.

Simple by design: it's just a few shell scripts.

Why / philosophy

• Who it's for
  • People who want to stay in their editor — (Neo)vim, Emacs — in the main tmux session
  • People who still read, write, and review the code themselves
  • People who'd rather have a small tool than a full-blown one just to manage agent sessions
• Who it's not for
  • People whose focus is the agent's chat interface rather than the code
  • People who want one tool for sessions across several AI providers — this one is Claude Code only
  • People who run a swarm of agents and don't read or write code themselves

Features

• A central picker (prefix + u) listing every running Claude agent — several in one project, and any running loose in an ordinary pane.
• Live status per agent — working / waiting / idle — read straight from claude agents --json, so you instantly see which need you. No setup.
• A live preview of each agent's screen right in the picker.
• Smart jump — selecting an agent switches your client to the window it was launched from, then resumes it in a popup over it.
• A launcher (prefix + y) that opens/attaches a Claude session for the current directory.
• Quick kill (ctrl-x) of a finished agent from the picker.
• **B