dmux-workflows

Orchestrate parallel AI agent panes in tmux sessions with merge steps.

1|Updated Mar 3, 2026
One-click install
npx skills add https://github.com/samymity/bridge-ventures-backend --skill dmux-workflows-samymity
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: dmux-workflows
Source: https://github.com/samymity/bridge-ventures-backend/tree/main/.claude/skills/dmux-workflows
Command: npx skills add https://github.com/samymity/bridge-ventures-backend --skill dmux-workflows-samymity

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Multi-agent development often stalls because agents must be manually started, coordinated, and merged, which wastes time and increases merge conflicts. dmux-workflows solves this by orchestrating parallel agent panes in a tmux session and providing repeatable patterns for splitting work safely.

Core Features & Use Cases

  • Parallel agent orchestration with dmux: Spin up multiple agent panes (each with its own session) and merge their outputs back into the main session for coordination.
  • Workflow patterns for common multi-agent tasks: Use targeted templates like Research + Implement, Multi-File Feature, Test + Fix Loop, Cross-Harness, and Code Review Pipeline to structure parallel work.
  • Conflict avoidance via git worktrees: Isolate overlapping file changes by running each pane in a separate worktree and merging branches when done.

Use cases include running independent feature workstreams (e.g., billing API + dashboard UI), speeding up debugging cycles (run tests in one pane, fix in another), or combining perspectives for security/performance/testing reviews across the same code area.

Quick Start

Start a tmux-based dmux session, then press n to open separate panes for your parallel agent prompts and press m to merge pane results 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 multiple AI agents in parallel to avoid merge conflicts?

Run parallel AI agents by orchestrating separate tmux panes within a dmux session, using git worktree isolation to prevent overlapping file changes and merging branches upon completion.

What is the best way to orchestrate a test and fix loop across multiple agents?

Orchestrate a test and fix loop by spinning up dedicated tmux panes for testing and fixing, allowing one agent to run tests while another applies fixes simultaneously before merging results.

Does dmux-workflows require git worktrees to manage multi-agent development?

Git worktrees are optional but recommended for multi-agent development to isolate overlapping file changes across panes and minimize merge conflicts during branch merging.

How do I merge outputs from parallel agent panes back into a main session?

Merge parallel agent pane outputs back into the main tmux session by pressing the m key, which consolidates the distributed workstreams into a single coordinated view.

Can I use tmux pane management to coordinate independent feature workstreams like billing and UI?

Yes, you can use tmux pane management to coordinate independent feature workstreams like billing APIs and dashboard UIs by running each in an isolated pane and merging results.

When should I avoid running parallel agent panes for code review pipelines?

Avoid running parallel agent panes for code review pipelines when tasks require deep sequential dependency chains, as coordinating merge steps for highly interdependent code changes increases overhead.