team

Orchestrate multiple Claude Code agents from a shared task list.

Updated Feb 15, 2026
One-click install
npx skills add https://github.com/leejaedus/pepcode --skill team-leejaedus
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: team
Source: https://github.com/leejaedus/pepcode/tree/main/skills/team
Command: npx skills add https://github.com/leejaedus/pepcode --skill team-leejaedus

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill orchestrates multiple Claude Code agents to work from a shared task list, enabling scalable collaboration and complex workflow coordination.

Core Features & Use Cases

  • Native team orchestration with a staged pipeline (plan, prd, exec, verify, fix) and per-stage agent routing
  • Hybrid MCP support for Gemini-based execution and large-context workflows
  • Robust state persistence, task creation, and teammate lifecycle management with resume, cancel, and audit capabilities
  • Safe shutdown, monitoring, and recovery to avoid duplicate teams and ensure deterministic progress

Quick Start

Invoke the team skill with a command such as /pepcode:team 3:executor 'refactor the auth module' to begin planning, spawning teammates, and executing the task.

Frequently Asked Questions about team

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

FAQPage Schema
How do I orchestrate multiple Claude Code agents to work from a shared task list?

You can orchestrate multiple Claude Code agents by invoking a team command to trigger a staged pipeline that plans, spawns teammates, and executes tasks from a shared task list with stateful tracking.

Can I resume or cancel a multi-agent workflow if an execution step fails?

Yes, the multi-agent orchestration workflow provides robust resume and cancel semantics, allowing you to manage the agent lifecycle and deterministically recover or halt tasks without losing progress.

Does multi-agent orchestration support hybrid MCP environments for large-context tasks?

Yes, the orchestration workflow supports hybrid MCP environments, enabling Gemini-based execution and handling large-context workflows alongside the standard Claude Code agents.

What is the staged pipeline for multi-agent task execution and verification?

The staged pipeline covers end-to-end team workflows by routing agents through distinct phases: planning, decomposing, spawning, executing, verifying, and fixing tasks to ensure production-grade progress.

How do I prevent duplicate agent teams during complex workflow coordination?

The orchestration workflow includes safe shutdown, monitoring, and recovery mechanisms to manage teammate lifecycles, avoid duplicate teams, and ensure deterministic progress across complex tasks.

What is the best way to coordinate a multi-agent team for refactoring a software module?

The best way to coordinate a multi-agent team for module refactoring is to invoke a single command that begins planning, automatically decomposes the module into subtasks, and spawns dedicated executor agents.