dmux-workflows

Orchestrate parallel AI agent sessions across development harnesses using tmux pane management.

2|Updated Apr 7, 2026
One-click install
npx skills add https://github.com/Zenobia000/ai-brainstorming --skill dmux-workflows-zenobia000
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: dmux-workflows
Source: https://github.com/Zenobia000/ai-brainstorming/tree/main/.claude/custom-rule%26skill/skills/dmux-workflows
Command: npx skills add https://github.com/Zenobia000/ai-brainstorming --skill dmux-workflows-zenobia000

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Manually coordinating multiple parallel AI agent sessions across different development tools is time-consuming and error-prone when splitting complex tasks into independent workstreams.

Core Features & Use Cases

  • Cross-harness orchestration: Manage parallel agent sessions across Claude Code, Codex, OpenCode, Cline, Gemini, and Qwen in a single tmux session.
  • Pre-built workflow patterns: Ready-to-use templates for common use cases including research + implementation splits, multi-file feature development, test-fix loops, and parallel code review.
  • Conflict-free collaboration: Git worktree integration guidance to avoid file conflicts between parallel agent workstreams.
  • Use case example: Split a complex e-commerce checkout feature build into parallel tracks for payment processing, inventory management, and UI components, then merge all results into a single cohesive implementation.

Quick Start

Use the dmux-workflows skill to split your current checkout feature build into parallel agent tracks for payment processing, inventory management, and UI components, then merge all results once each track is complete.

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 without manual coordination?

You can orchestrate parallel AI agent sessions across multiple development harnesses using tmux pane management to eliminate manual coordination overhead. Pre-built workflow patterns split complex software development, research, and code review tasks into independent parallel workstreams.

Can I manage Claude Code and Codex sessions in a single tmux window?

Yes, cross-harness orchestration manages parallel agent sessions across Claude Code, Codex, OpenCode, Cline, Gemini, and Qwen in a single tmux session. This allows you to monitor and coordinate multiple AI tools simultaneously without switching contexts.

How do I prevent file conflicts when running parallel AI coding agents?

Git worktree integration guidance provides conflict avoidance best practices for parallel agent workstreams. By isolating each agent in a separate worktree, multiple agents can edit code simultaneously without overwriting each other's changes.

What are the best workflow patterns for splitting complex feature development?

Ready-to-use templates cover research and implementation splits, multi-file feature development, test-fix loops, and parallel code review. For example, you can split an e-commerce checkout feature into parallel tracks for payment processing, inventory, and UI, then merge results.

Does dmux-workflows support AI agents beyond Claude Code?

Yes, cross-harness orchestration supports Claude Code, Codex, OpenCode, Cline, Gemini, and Qwen. It provides pre-built workflow patterns and conflict avoidance best practices applicable to these and other agent tools for complex software development tasks.

When should I use parallel agent workflows for software development?

Apply parallel agent workflows to complex software development, research, and code review tasks that can be split into independent workstreams. If a task has independent tracks like payment processing and UI components, parallel orchestration merges them into a cohesive implementation.