ralph-parallel

Orchestrate parallel Claude task execution with git worktrees and tmux windows.

2|Updated Dec 28, 2024
One-click install
npx skills add https://github.com/shonenm/dotfiles --skill ralph-parallel
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ralph-parallel
Source: https://github.com/shonenm/dotfiles/tree/main/common/claude/.claude/skills/ralph-parallel
Command: npx skills add https://github.com/shonenm/dotfiles --skill ralph-parallel

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires Bash, Read, Write, Glob, Grep, Task, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill orchestrates parallel execution of tasks defined in a state file or task list, managing dependencies and providing visibility into each task's progress.

Core Features & Use Cases

  • Dependency Management: Analyzes task graphs to ensure tasks run in the correct order.
  • Parallel Execution: Runs up to four independent Claude processes concurrently.
  • Visibility: Utilizes git worktrees and tmux windows for real-time monitoring of each task.
  • Use Case: For a complex feature development, break down the work into smaller, independent tasks. This Skill will manage their execution, allowing multiple AI agents to work on different parts simultaneously, with you able to watch their progress in separate terminal windows.

Quick Start

Execute the ralph-parallel skill to process tasks from the default state file.

Frequently Asked Questions about ralph-parallel

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

FAQPage Schema
How do I run parallel Claude processes for independent tasks?

You can run parallel Claude processes by defining tasks in a state file or task list. The Skill orchestrates up to four concurrent Claude processes, analyzing dependencies to ensure correct execution order.

How does dependency analysis work for parallel task execution?

Dependency analysis for parallel task execution works by parsing a task graph from your state file to determine execution order. This ensures dependent tasks wait, while independent tasks run concurrently in isolated environments.

Do I need git worktree and tmux to monitor parallel task execution?

Yes, git worktree and tmux are required to monitor parallel task execution. Each independent Claude process runs in an isolated git worktree and tmux window, allowing real-time observation of task status and progress.

What is the best way to manage multiple AI agents working on different parts of a feature?

The best way to manage multiple AI agents is breaking complex feature development into smaller independent tasks. The Skill manages their execution, allowing agents to work simultaneously with progress visible in separate terminal windows.

Can I use Bash and Task tools to orchestrate task execution from a state file?

Yes, Bash and Task tools are supported alongside Read, Write, Glob, and Grep for orchestrating task execution. These dependencies enable the Skill to process state files, analyze dependencies, and manage parallel Claude processes.

Why does parallel task execution limit concurrent processes to four?

Parallel task execution limits concurrent processes to four to balance resource usage and performance. This constraint allows independent Claude processes to run simultaneously without overwhelming system resources while maintaining real-time visibility.