dmux-workflows

Coordinate parallel AI agent sessions across multiple development harnesses.

Updated May 14, 2026
One-click install
npx skills add https://github.com/WUIX69/e-lock --skill dmux-workflows-wuix69
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: dmux-workflows
Source: https://github.com/WUIX69/e-lock/tree/main/.agent/.agents/skills/dmux-workflows
Command: npx skills add https://github.com/WUIX69/e-lock --skill dmux-workflows-wuix69

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you coordinate multiple AI agent sessions at once so you can split large tasks into smaller parallel tracks instead of handling everything in a single chat.

Core Features & Use Cases

  • Parallel orchestration: Launch separate agent panes for research, implementation, testing, and review.
  • Cross-harness workflows: Coordinate sessions across Claude Code, Codex, OpenCode, Cline, Gemini, and Qwen.
  • Safe merging patterns: Use clear task boundaries and merge results back into a unified workflow after each pane finishes.
  • Use case: When a feature spans several files or needs independent review passes, you can assign each part to a different agent and combine the outputs efficiently.

Quick Start

Use the dmux-workflows skill to split my task into parallel agent panes, assign each pane an independent subtask, and merge the results into one coordinated plan.

Frequently Asked Questions about dmux-workflows

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

FAQPage Schema
How do I coordinate parallel AI agent sessions for complex development work?

You can coordinate parallel AI agent sessions by using dmux to split large tasks into separate panes for research, implementation, testing, and review, then merging the independent outputs back into a unified workflow.

Can I orchestrate multi-agent workflows across different coding harnesses like Claude Code and Codex?

Yes, multi-agent orchestration supports cross-harness workflows, allowing you to coordinate parallel sessions across Claude Code, Codex, OpenCode, Cline, Gemini, and Qwen for unified task execution.

What is the best way to isolate parallel AI coding tasks using git worktree?

Git worktree isolation allows you to assign independent subtasks to different AI agent panes safely, ensuring that parallel implementation tracks remain separate until you merge results back together.

How do I merge independent outputs from multiple AI agent panes into one plan?

You merge independent outputs by establishing clear task boundaries for each agent pane, allowing them to finish their subtasks, and then applying structured merge-back patterns to combine the results.

When do I need parallel agent orchestration for software engineering tasks?

You need parallel agent orchestration when a feature spans several files or requires independent review passes, allowing you to divide the work into smaller parallel tracks instead of handling everything in a single chat.

Does dmux-workflows require tmux for pane management during multi-agent tasks?

Yes, dmux-based pane management relies on tmux to launch and manage separate agent panes, enabling you to monitor and coordinate multiple parallel AI sessions simultaneously.