multi-agent

Manage multiple AI coding agents via PTY mode and git worktrees.

Updated Feb 18, 2026
One-click install
npx skills add https://github.com/ollieb89/openrepo --skill multi-agent-ollieb89
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: multi-agent
Source: https://github.com/ollieb89/openrepo/tree/main/skills/multi-agent
Command: npx skills add https://github.com/ollieb89/openrepo --skill multi-agent-ollieb89

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill streamlines the execution of multiple AI agents, from simple one-off tasks to complex, interdependent workflows, ensuring efficient and reliable operation.

Core Features & Use Cases

  • Individual Agent Execution: Run coding agents like Codex, Claude, and Gemini with PTY mode for interactive terminal applications.
  • Multi-Agent Orchestration: Coordinate agents with explicit dependencies, manage tasks across worktrees, and implement self-healing monitoring for robust workflows.
  • Use Case: Automate the process of fixing bugs across multiple GitHub issues by spawning individual agents for each task, managing their dependencies, and monitoring their progress until all are resolved.

Quick Start

Run the 'codex' agent with PTY enabled in the '~/project' directory to build a snake game.

Frequently Asked Questions about multi-agent

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

FAQPage Schema
How do I coordinate multiple AI coding agents to work on interdependent tasks?

You can orchestrate multiple AI coding agents by managing explicit dependencies, executing tasks in parallel across git worktrees, and implementing self-healing monitoring to ensure robust and reliable interdependent workflows.

Can I run interactive terminal applications with AI agents like Claude and Gemini?

Yes, coding agents like Codex, Claude, and Gemini can run interactive terminal applications using PTY mode, providing a bash-first approach that supports programmatic control via background processes for reliable execution.

What is the best way to automate fixing multiple GitHub issues using AI agents?

Spawning individual coding agents for each GitHub issue allows you to automate bug fixes by managing task dependencies, coordinating parallel execution, and monitoring progress until all issues are fully resolved.

Does multi-agent orchestration support parallel execution across different worktrees?

Yes, the multi-agent orchestration approach supports parallel execution by utilizing git worktrees and tmux, allowing multiple AI coding agents to operate concurrently while maintaining isolated working directories.

How do I monitor and manage active sessions for running AI agents?

Session management and monitoring tools allow you to track active AI agent sessions, control background processes, and implement self-healing workflows to maintain reliable operation across complex multi-agent tasks.

What are the limitations of using a bash-first approach for AI agent orchestration?

While the bash-first approach with PTY mode enables interactive terminal applications, it requires careful management of background processes and terminal interactions when coordinating multiple agents with explicit dependencies.