openwong2kim

openwong2kim / wmux

Public

AI Agent Terminal for Windows — Run Claude Code, Codex, Gemini CLI side by side with split terminals, browser automation via CDP, MCP integration, session persistence, and smart notifications

10
5
100% credibility
Found Mar 25, 2026 at 10 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

wmux is a Windows app that lets users split terminals and browsers into organized workspaces for running multiple AI coding agents simultaneously with built-in notifications and persistence.

How It Works

1
🔍 Discover wmux

You're frustrated running one AI coding helper at a time on Windows and hear about wmux, a handy screen splitter just for that.

2
📥 Grab it quickly

Download the simple installer or run one easy command to get wmux on your computer in seconds.

3
🚀 Open your hub

Launch wmux and see a fresh terminal ready to go, with easy shortcuts to split screens.

4
✂️ Split and organize

Use simple keys to divide your screen into multiple terminals or add a browser panel side-by-side.

5
🤖 Team up with AI

Your AI coding friends like Claude automatically connect and control browsers or terminals without extra setup.

6
🔔 Stay in the loop

Get pop-up alerts when tasks finish or spot risky commands before they run.

🎉 Work flows smoothly

Run multiple AI agents at once, sessions stick around even after restarts, and everything feels fast and reliable.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 10 to 10 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is wmux?

wmux is a TypeScript-based Electron app that brings tmux-style terminal multiplexing to Windows, letting you run AI agent CLIs like Claude Code, Gemini, or Codex side-by-side in split panes. It solves the pain of juggling multiple agent terminals on Windows—where tmux doesn't exist natively—by adding browser automation via Chrome DevTools Protocol, MCP integration for Claude, session persistence across reboots, and CLI commands like `wmux send "npm test"` or `wmux browser navigate https://example.com`. Developers get a single window for multi-agent workflows with smart notifications on task completion.

Why is it gaining traction?

Unlike basic terminals, wmux detects agents like Cursor or GitHub Copilot CLI, warns on risky commands like `git push --force`, and lets agents control browsers directly—no more "agent terminals are read only" issues in Cursor. Its daemon keeps sessions alive post-crash or reboot, with 999K-line scrollback, while MCP tools like `terminal_read` and `browser_snapshot` make it plug-and-play for chatgpt agent terminal or copilot agent in terminal setups. Early adopters praise the Windows-first focus, filling a gap left by macOS tools like cmux.

Who should use this?

Windows devs running agent github claude, agent github copilot, or agent github openai in parallel for full-stack tasks—Claude on backend, Gemini on tests, Codex on frontend. Ideal for terminal agent job runners tired of tab-switching, or those debugging cursor agent terminal not working errors. Skip if you're on Linux with tmux or don't need browser CDP for agent github action automation.

Verdict

Try wmux if you're on Windows building AI-assisted repos—its MCP and persistence shine for multi-agent flows, but with 10 stars and 1.0% credibility score, it's early alpha: solid README, but expect bugs and light tests. Production? Wait for 100+ stars.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.