cmux

Manage multiple Claude Code agent sessions across named workspaces in a terminal.

2|Updated Feb 23, 2026

Search Tags:#cli #notifications #terminal #workspaces #orchestrator #cmux

Ready to run

npx skills add https://github.com/jonmumm/skills --skill cmux-jonmumm

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: cmux
Source: https://github.com/jonmumm/skills/tree/main/cmux
Command: npx skills add https://github.com/jonmumm/skills --skill cmux-jonmumm

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

cmux provides centralized management of multiple Claude Code agent sessions in parallel terminals, enabling coordinated workflows and easier monitoring.

Core Features & Use Cases

Create, switch, and monitor named workspaces to run several agents concurrently.
Send commands between workspaces, split surfaces, and manage notifications and status updates.
Use for /swarm or /nightshift style orchestration where terminal-level control and visibility are required.

Quick Start

Spawn a new workspace and start an agent to begin orchestrating parallel sessions.

Frequently Asked Questions about cmux

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I manage multiple Claude Code agent sessions in parallel terminals?▼

To run Claude Code agents in parallel, use a terminal orchestrator to create named workspaces for each agent. This approach allows you to manage concurrent sessions within a single terminal, providing centralized control and monitoring for parallel workflows.

What is terminal workspace orchestration for AI agents?▼

Terminal workspace orchestration coordinates multiple AI agent sessions by creating named workspaces within a single terminal. This enables concurrent agent execution, inter-workspace communication, and centralized status monitoring for complex parallel workflows.

Do I need a macOS-compatible terminal to run parallel agent workspaces?▼

Yes, you need a macOS-compatible terminal environment to use this workspace orchestrator. The CLI relies on terminal-level control and visibility features specific to macOS to create, switch, and monitor parallel agent sessions effectively.

Can I send commands between workspaces when running parallel AI agents?▼

Yes, the orchestrator allows sending commands between workspaces while running parallel AI agents. It supports splitting terminal surfaces and managing notifications across workspaces to maintain coordinated workflows and inter-agent communication.

What are the limitations of using a CLI orchestrator for parallel agent sessions?▼

The primary limitation is platform dependency; the orchestrator requires a macOS-compatible terminal environment to function. It is designed specifically for terminal-level control and visibility, meaning it does not support non-CLI based agent workflows.