Where are sub-agent definitions stored?

Sub-agent definitions are authored as markdown files typically located under the agents/ directory of your project with specific frontmatter metadata.

What are delegation patterns in this context?

Delegation patterns define the communication protocol and authority boundaries between parent agents and sub-agents, ensuring tasks are passed and executed correctly.

What is a Claude sub-agent?

A sub-agent is a specialized AI instance within Claude Code designed to handle specific tasks or roles within a larger workflow or team structure, defined via markdown configuration.

How does this skill manage sub-agent permissions?

It implements a least-privilege strategy, allowing you to define exactly which tools and models each sub-agent can access to ensure security and operational focus.

Can I validate my agent configurations?

Yes, the skill includes specific scripts to validate frontmatter and the structural integrity of your agent markdown files before deployment.

Sub-Agent Implementation

Name: Sub-Agent Implementation
Author: yu-iskw

byyu-iskw

•

数据科学与机器学习

Architects and validates specialized Claude sub-agents with precise role boundaries and tool permissions.

This skill empowers developers to build sophisticated multi-agent systems within the Claude Code ecosystem by defining clear hierarchies and delegation patterns. It provides a structured framework for creating sub-agents with specific model configurations, frontmatter requirements, and least-privilege tool access, ensuring each agent operates effectively within its designated scope. Whether you are building a complex software development team or a simple automated workflow, this skill streamlines the process of agent orchestration and validation through standardized templates and automated checking scripts.

主要功能

01Least-privilege tool and permission configuration

02Built-in scripts for frontmatter checking and agent file validation

03Standardized sub-agent frontmatter creation and validation

04Comprehensive templates for various agent types (Read-only, Skill-based, Hook-based)

05Explicit role boundary and delegation pattern definition

063 GitHub stars

使用场景

01Setting up automated validation workflows to ensure sub-agent configuration consistency

02Creating specialized read-only agents for codebase analysis and documentation

03Designing a 3-tier software development team with planning, orchestration, and execution layers

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add yu-iskw/software-development-team-plugin implement-sub-agents

For use in Claude.ai and ChatGPT

主要功能

01Least-privilege tool and permission configuration

02Built-in scripts for frontmatter checking and agent file validation

03Standardized sub-agent frontmatter creation and validation

04Comprehensive templates for various agent types (Read-only, Skill-based, Hook-based)

05Explicit role boundary and delegation pattern definition

063 GitHub stars

使用场景

01Setting up automated validation workflows to ensure sub-agent configuration consistency

02Creating specialized read-only agents for codebase analysis and documentation

03Designing a 3-tier software development team with planning, orchestration, and execution layers

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add yu-iskw/software-development-team-plugin implement-sub-agents

For use in Claude.ai and ChatGPT