Back to browse
GitHub Repository

A tmux environment built for running coding agents in parallel — with a persistent sidebar that shows every session, what's running, and what needs your attention.

56 starsTypeScript

Jmux – tmux-based development environment for humans and coding agents

by jarredkenny·Apr 5, 2026·10 points·10 comments

AI Analysis

●●●BangerSolve My ProblemNiche GemSlick

0.3MB tmux layer beats 100MB Electron orchestrators — keeps your existing workflow.

Strengths
  • Orange attention flags show which agents finished and need review — no guessing.
  • Sessions auto-grouped by parent directory with git branch display in sidebar.
  • Works over SSH and containers — anywhere tmux runs, not just macOS.
Weaknesses
  • Requires tmux familiarity — non-tmux users won't adopt this.
  • Bun runtime dependency may friction for npm/yarn-only environments.
Target Audience

Terminal developers running multiple AI coding agents simultaneously

Similar To

Conductor · cmux · Tmuxinator

Post Description

I've been a tmux user for years. When I started running 5-10 Claude Code sessions in parallel, I tried the tools that are out there: Conductor, cmux, the GUI orchestrators. None of them felt right. They either wanted me to leave tmux entirely for a 100MB+ Electron app with its own editor and Git workflow, or they were thin wrappers that didn't solve the actual problem: I need to parallelize my entire development environment, agents, editors, servers, logs, and keep track of all of it.

So I built jmux. It turns tmux into a parallel development environment. Every session, every agent, every running process, visible and navigable from one terminal. When Claude Code completes a response, an orange flag appears on that session. Switch to it, review the work, move on. No context switching to a different app. No learning a new workflow. It's still tmux underneath.

The key decision was to build on tmux instead of replacing it. My ~/.tmux.conf still works. My neovim setup still works. My keybindings, plugins, prefix key, etc, all of it carries over. jmux just makes it all work at scale.

What it does:

- Run Claude Code, Codex, aider. Any agent, directly, in parallel

- Attention flags. Know the moment an agent finishes without watching every pane

- Instant switching between sessions (Ctrl-Shift-Up/Down, no prefix key)

- Auto-groups sessions by project and git branch

- Works with your existing tmux config, editor, and tools

- No SDK, no API wrappers. First-party tools, no middleman.

What it doesn't do:

- No built-in editor (I use neovim btw)

- No built-in Git (use yours)

- No agent protocol/wrapper (run them directly in the terminal)

~1800 lines of TypeScript, ~0.3 MB installed. MIT licensed.

Landing page: https://jmux.build

Github: https://github.com/jarredkenny/jmux

Similar Projects