dmux-workflows

Orchestrate parallel AI agent sessions across tmux panes with dmux.

Updated Jun 22, 2026
One-click install
npx skills add https://github.com/betaTrident/manta --skill dmux-workflows-betatrident
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: dmux-workflows
Source: https://github.com/betaTrident/manta/tree/main/.agents/skills/dmux-workflows
Command: npx skills add https://github.com/betaTrident/manta --skill dmux-workflows-betatrident

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires dmux.

What problem does it solve?

This Skill solves the bottleneck of sequential task execution by enabling multi-agent orchestration, allowing you to run multiple AI sessions in parallel to divide and conquer complex development tasks.

Core Features & Use Cases

  • Parallel Agent Management: Manage multiple AI agent panes within a single terminal session using tmux.
  • Workflow Patterns: Supports specialized patterns like Research + Implement, Multi-File Feature development, and Test + Fix loops.
  • Use Case: When building a new feature, use this skill to simultaneously run one agent for database schema creation, another for API endpoint development, and a third for UI component generation, then merge the results.

Quick Start

Activate the dmux-workflows skill to initialize a new multi-pane session and begin parallelizing your current development tasks.

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 for concurrent development workflows?

You can run parallel AI agent sessions by orchestrating multiple tmux panes to execute concurrent development workflows, allowing you to divide and conquer complex software engineering tasks simultaneously.

What is multi-agent orchestration and when should I use it for software engineering?

Multi-agent orchestration manages multiple AI sessions in parallel to divide and conquer complex development tasks. It is needed when sequential task execution creates a bottleneck, such as building database schemas, APIs, and UI components simultaneously.

Do I need tmux to manage multiple AI panes in a single terminal?

Yes, you need a tmux-compatible environment and the dmux npm package. These facilitate creating multiple panes within a single terminal session and merging the output from your concurrent AI agents.

How do I use divide-and-conquer strategies across different agent harnesses?

You apply divide-and-conquer strategies by assigning specialized workflow patterns like Research + Implement or Test + Fix loops to separate managed AI panes, then merging the results of the concurrent execution.

What is the best way to develop multi-file features using parallel AI agents?

The best way to develop multi-file features is using parallel agent management to run specialized sessions simultaneously, such as having one agent handle database schema creation while another generates API endpoints.

Can I merge output from concurrent tmux panes running separate AI tasks?

Yes, the dmux npm package facilitates pane creation and output merging, allowing you to combine the results of concurrent AI agent tasks running across separate tmux panes into a unified workflow.