taskmaestro

Orchestrate multiple Claude Code instances across tmux panes with isolated git worktrees.

31|3|Updated Dec 12, 2025
One-click install
npx skills add https://github.com/JeremyDev87/codingbuddy --skill taskmaestro
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: taskmaestro
Source: https://github.com/JeremyDev87/codingbuddy/tree/main/.claude/skills/taskmaestro
Command: npx skills add https://github.com/JeremyDev87/codingbuddy --skill taskmaestro

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Coordinating multiple Claude Code instances across tmux panes with isolated git worktrees to manage concurrent development tasks without context switching.

Core Features & Use Cases

  • Orchestrates parallel Claude Code sessions across multiple tmux panes.
  • Enforces per-pane git worktree isolation and deterministic task execution.
  • Provides preflight checks, RESULT.json signaling, and optional review automation to streamline collaboration.

Quick Start

Launch TaskMaestro in your tmux session and begin assigning tasks to worker panes using the assign subcommand.

Frequently Asked Questions about taskmaestro

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

FAQPage Schema
How do I run parallel Claude Code tasks in tmux without context switching?

Orchestrating parallel Claude Code tasks in tmux involves distributing multiple instances across separate panes with isolated git worktrees, enabling concurrent development without context switching. The Skill manages task assignment, monitors status, and consolidates pane-specific workflow control.

What is git worktree isolation for concurrent Claude Code development?

Git worktree isolation assigns a separate working directory to each tmux pane, ensuring deterministic task execution for Claude Code. This prevents file conflicts when multiple parallel instances modify code simultaneously during coordinated workflow orchestration.

Does TaskMaestro require tmux to orchestrate Claude Code sessions?

Tmux is required as the environment to orchestrate multiple Claude Code sessions across panes. The Skill manages pane-specific workflows, preflight checks, and RESULT.json signaling within a single tmux session to coordinate parallel development tasks.

How do I assign tasks to specific tmux panes for parallel development?

You assign tasks to specific tmux panes using the assign subcommand. This directs individual Claude Code instances to execute isolated worktree tasks while the orchestrator monitors execution status and collects RESULT.json output signals from each worker.

Can I automate code reviews after parallel Claude Code tasks finish?

Optional review automation is supported after parallel tasks finish. The orchestration workflow triggers review checks once per-pane task execution completes and RESULT.json signals are collected from the tmux worker panes, streamlining collaborative code validation.

What are the limitations of running multiple Claude Code instances in tmux?

Running multiple Claude Code instances in tmux requires careful resource management, as each isolated worktree pane consumes compute independently. Preflight checks validate environment readiness, but complex dependencies between parallel tasks may still create coordination bottlenecks during execution.