dmux-workflows

Orchestrate parallel AI agent sessions across multiple harnesses using dmux.

3|2|Updated Mar 8, 2026
One-click install
npx skills add https://github.com/agentmatters/mullai-bot --skill dmux-workflows-agentmatters
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: dmux-workflows
Source: https://github.com/agentmatters/mullai-bot/tree/main/src/Mullai.Skills/Skills/claude-code-everything/dmux-workflows
Command: npx skills add https://github.com/agentmatters/mullai-bot --skill dmux-workflows-agentmatters

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Orchestrate parallel AI agent sessions using dmux, a tmux-based pane manager for harnesses.

Core Features & Use Cases

  • Split workflows across multiple panes for parallel experimentation across Claude Code, Codex, OpenCode, and other harnesses.
  • Quickly create, merge, and manage agent panes to run parallel tasks and track results.
  • Use cases include cross-harness development, rapid prototyping, and coordinated multi-pane experiments.

Quick Start

Start a dmux session and create agent panes to run parallel workflows.

Frequently Asked Questions about dmux-workflows

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

FAQPage Schema
How do I run parallel AI agent sessions in tmux panes?

You can run parallel AI agent sessions by using dmux, a tmux-based pane manager that creates and coordinates multiple harness panes for concurrent workflow execution.

Can I coordinate Claude Code and Codex sessions in the same tmux workspace?

Yes, dmux supports cross-harness development by splitting workflows across multiple panes to run Claude Code, Codex, and OpenCode sessions in parallel within a single tmux workspace.

What do I need to set up before orchestrating parallel agent workflows with dmux?

You need a terminal with tmux installed and the dmux installation from its repository to start orchestrating parallel agent sessions and managing harness panes.

How do I create and merge agent panes for parallel experimentation?

You create new agent panes using the n command and merge panes using m within a dmux session to manage parallel tasks and track experiment results across harnesses.

What is the best way to manage multiple AI coding agents running at the same time?

Using a tmux-based pane manager like dmux is an effective approach to orchestrate parallel AI agent sessions, allowing you to quickly create, merge, and track work across multiple harness panes.

Are there limitations when using tmux for parallel AI agent orchestration?

Parallel AI agent orchestration with dmux is limited to terminal environments supporting tmux, requiring the specific dmux installation to manage panes and coordinate across supported harnesses.