team

Coordinate multiple autonomous agents in tmux panes with shared state and mailbox communication.

3|Updated Apr 7, 2026
One-click install
npx skills add https://github.com/Tienching/oh-my-codebuddy --skill team-tienching
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: team
Source: https://github.com/Tienching/oh-my-codebuddy/tree/main/.codex/skills/team
Command: npx skills add https://github.com/Tienching/oh-my-codebuddy --skill team-tienching

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The team skill enables reliable, durable coordination of multiple automated workers via a tmux-based runtime, providing shared task state, mailbox-driven communication, and a scalable execution model for complex pipelines.

Core Features & Use Cases

  • Durable parallel execution: launches real worker sessions (Codex/Claude) in split panes and coordinates them through a central state store.
  • Shared task state & mailbox: uses .omb/state/team/... files and omb team api for inter-process coordination and lifecycle management.
  • Worktrees & lifecycle control: supports multiple worktrees and explicit lifecycle hooks for long-running team workflows across leader and workers.
  • Use cases include long-running feature work, parallel evaluation, and coordinated validation across multiple agents.

Quick Start

Start a team run by providing a clear task to omb team, for example: omb team 2:executor "analyze feature X and report findings".

Frequently Asked Questions about team

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

FAQPage Schema
How do I coordinate multiple autonomous agents to run tasks in parallel?

To coordinate parallel agents, you can use a tmux-driven team to launch real worker sessions in split panes, sharing task state and mailbox communication to execute long-running pipelines reliably.

How does mailbox communication work for inter-process coordination between parallel workers?

Mailbox communication enables inter-process coordination by using shared state files under a central root and CLI interop, allowing leader and worker panes to exchange structured messages and manage task lifecycles.

Can I use tmux split panes to manage long-running feature work across multiple agents?

Yes, tmux split panes support long-running feature work by launching real worker sessions that coordinate through a central state store, providing durable execution and explicit lifecycle control across leader and worker panes.

What's the best way to orchestrate parallel evaluation across multiple automated workers?

Orchestrating parallel evaluation is best handled by launching a tmux-driven team that coordinates autonomous workers through shared task state and mailbox semantics, ensuring structured task lifecycles and durable worker coordination.

Do I need a specific state directory structure to manage parallel agent execution?

Yes, parallel agent execution requires a unified team state root directory to store shared state files, enabling mailbox-driven communication and lifecycle management across leader and worker panes.

When should I avoid using a tmux-based runtime for parallel agent orchestration?

You should avoid tmux-based parallel orchestration for short-lived, single-agent tasks that do not require durable worker coordination, persistent state management, or structured task lifecycles across multiple panes.