subagent-custom-agent

Create custom .agent.md files for GitHub Copilot subagents.

Updated Mar 6, 2026
One-click install
npx skills add https://github.com/Zoimback/copilot-dev-core --skill subagent-custom-agent
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: subagent-custom-agent
Source: https://github.com/Zoimback/copilot-dev-core/tree/main/.github/skills/subagent-custom-agent
Command: npx skills add https://github.com/Zoimback/copilot-dev-core --skill subagent-custom-agent

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill streamlines the creation of specialized AI agents, enabling users to define custom behaviors and toolsets for specific tasks within the GitHub Copilot environment.

Core Features & Use Cases

  • Define Specialized Agents: Create agents with unique roles, instructions, and tool access.
  • Configure Subagents: Build worker agents or reusable components for delegation.
  • Control Invocation: Manage how agents can be triggered by users or other AI agents.
  • Use Case: You need an agent that can only read code and search for security vulnerabilities. This skill allows you to define such an agent with specific tools and constraints.

Quick Start

Use the subagent-custom-agent skill to create a new agent named 'code-linter' that can only read files and search the codebase.

Frequently Asked Questions about subagent-custom-agent

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

FAQPage Schema
How do I create custom AI agents for GitHub Copilot?

To create custom AI agents for GitHub Copilot, you define specialized roles, instructions, and tool access in .agent.md files, which are then used as subagents within the environment.

Can I configure subagents to restrict tool access for specific tasks?

Yes, you can configure subagents to restrict tool access, allowing you to define specialized agents that can only read files and search the codebase for specific tasks like security analysis.

What is the best way to manage how AI agents are triggered in Copilot?

The best way to manage AI agent invocation is by defining invocation control rules within your .agent.md files, which manages how agents are triggered by users or other AI workers.

How does defining worker agents help with codebase tasks?

Defining worker agents allows you to build reusable components for delegation, enabling specialized agents to handle specific codebase tasks with assigned models and constrained instructions.

Do I need any dependencies to build specialized agents with .agent.md files?

No dependencies are required to build specialized agents, as the skill operates independently to define .agent.md files for configuring models, tools, and invocation rules within GitHub Copilot.