dmux-workflows

Coordinate parallel AI agent sessions across multiple harnesses using dmux.

Updated Apr 19, 2026
One-click install
npx skills add https://github.com/rivaldiekaptrrr/Track-app --skill dmux-workflows-rivaldiekaptrrr
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: dmux-workflows
Source: https://github.com/rivaldiekaptrrr/Track-app/tree/main/.agent/.agents/skills/dmux-workflows
Command: npx skills add https://github.com/rivaldiekaptrrr/Track-app --skill dmux-workflows-rivaldiekaptrrr

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Orchestrate and manage multiple AI agent sessions in parallel using dmux, reducing manual coordination overhead across tools and harnesses.

Core Features & Use Cases

  • dmux-based pane management for parallel agent sessions across Claude Code, Codex, OpenCode, and other harnesses.
  • Create, split, and merge agent panes to run tasks in parallel while preserving context.
  • Use cases include research + implementation pipelines, multi-file feature work, and cross-harness collaboration.

Quick Start

Start a dmux session and create multiple agent panes, then merge their outputs back into the main session.

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?

You can run parallel AI agent sessions in tmux by using dmux to create, split, and manage multiple panes. This allows you to coordinate tasks across different agent harnesses like Claude Code and Codex while preserving context.

What is dmux used for in multi-agent workflows?

Dmux is used for orchestrating and managing multiple AI agent sessions in parallel within tmux. It reduces manual coordination overhead by providing pane-management commands to divide work among different agent tools and merge their outputs.

Can I use dmux to coordinate Claude Code and Codex agents together?

Yes, dmux supports cross-harness collaboration, allowing you to coordinate Claude Code, Codex, and OpenCode agents simultaneously. You can create separate panes for each harness and merge their results back into a main session.

How do I merge outputs from multiple AI agents in a tmux session?

To merge outputs from multiple AI agents, use dmux pane-management commands to consolidate results from individual agent panes back into your main tmux session. This preserves the context of each parallel task during the merge.

When should I use parallel agent sessions instead of a single AI agent?

Use parallel agent sessions when dividing complex work like research and implementation pipelines or multi-file feature development. Running multiple harnesses simultaneously reduces bottlenecks compared to sequential single-agent execution.

Do I need tmux installed to manage parallel AI workflows with dmux?

Yes, tmux is required because dmux relies on a tmux-based workflow to provide the pane-management commands for creating, managing, and merging parallel AI agent outputs.