subagent-collaboration-workflow

Coordinate multi-agent workflows with role-based delegation and centralized verification.

28|2|Updated May 4, 2026
One-click install
npx skills add https://github.com/xiaohei-info/oh-my-agent-skills --skill subagent-collaboration-workflow
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: subagent-collaboration-workflow
Source: https://github.com/xiaohei-info/oh-my-agent-skills/tree/main/skills/multi-agent-control/subagent-collaboration-workflow
Command: npx skills add https://github.com/xiaohei-info/oh-my-agent-skills --skill subagent-collaboration-workflow

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This skill solves the problem of messy, uncoordinated multi-agent delegation where subagents drift, overlap, or fail to produce verifiable results.

Core Features & Use Cases

  • Structured Role Delegation: Implements a clear hierarchy including Planner, Investigator, Implementer, and Reviewer roles.
  • Centralized Verification: Ensures all subagent work is independently verified by the controller before claiming success.
  • Use Case: Use this when tackling complex codebase refactors that require a plan, parallel investigation of dependencies, bounded implementation, and multi-stage quality review.

Quick Start

Use the subagent collaboration workflow skill to decompose the current task into planner investigator and implementer roles with independent verification.

Frequently Asked Questions about subagent-collaboration-workflow

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

FAQPage Schema
How do I coordinate multi-agent workflows to prevent subagents from overlapping or drifting during complex refactoring?

Multi-agent workflows are coordinated by enforcing structured role-based delegation and centralized verification. This controller-side oversight ensures subagents remain bounded and independently verified before claiming success.

What is the best way to decompose multi-step software engineering tasks for parallel investigation?

Multi-step software engineering tasks are decomposed by systematically partitioning them into Planner, Investigator, and Implementer roles. This structured task partitioning enables parallel investigation of dependencies and bounded implementation.

How does centralized verification work for subagent delegation in codebase refactoring?

Centralized verification requires the controller to independently review all subagent work before claiming success. This ensures codebase refactoring results are rigorously checked through multi-stage quality review and independent spec review.

When do I need role-based delegation for multi-agent software engineering tasks?

Role-based delegation is needed for complex multi-step software engineering tasks requiring decomposition, parallel investigation, and rigorous quality control. It prevents uncoordinated multi-agent delegation where subagents drift or fail to produce verifiable results.

Can I use this subagent collaboration workflow for tasks without independent spec review requirements?

This subagent collaboration workflow is designed for tasks requiring controller-side oversight, independent spec review, and systematic task partitioning. It is not suited for simple tasks lacking multi-stage quality control or parallel investigation needs.