create-agent

Scaffold GitHub Copilot custom agent profiles under .github/agents with YAML frontmatter.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/kirillshilin/instructions --skill create-agent-kirillshilin
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: create-agent
Source: https://github.com/kirillshilin/instructions/tree/main/obsolete/dotgithub/skills/create-agent
Command: npx skills add https://github.com/kirillshilin/instructions --skill create-agent-kirillshilin

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps teams scaffold a new GitHub Copilot custom agent profile following repository conventions and the official Copilot guidelines, ensuring consistency and readiness for deployment.

Core Features & Use Cases

  • Structured agent template: Generates a complete profile with frontmatter and a prompt body that adheres to the repository's conventions.
  • Guided workflow: Outlines a step-by-step process to clarify intent, choose a filename, write frontmatter, craft the prompt, validate, and save under the correct path.
  • Scalability & governance: Enables repeatable agent creation for multiple projects while enforcing naming and tool constraints.

Quick Start

Create a new agent file at .github/agents/create-agent.agent.md by following the steps described above.

Frequently Asked Questions about create-agent

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

FAQPage Schema
How do I create a custom GitHub Copilot agent profile for my repository?

To create a GitHub Copilot agent, scaffold a profile file under .github/agents using valid YAML frontmatter and a structured prompt body. The process includes naming, filename derivation, frontmatter validation, and prompt construction to meet repository conventions.

What is the correct file path and format for defining a Copilot agent?

A Copilot agent is defined as a Markdown file saved under the .github/agents directory. The file must contain valid YAML frontmatter for configuration and a descriptive prompt body that outlines the agent's behavior and tool constraints.

How do I ensure my Copilot agent follows repository conventions and governance?

To ensure your Copilot agent follows repository conventions, use a guided scaffolding workflow that enforces consistent naming, filename derivation, and YAML frontmatter validation. This approach provides repeatable agent creation and enforces tool constraints across multiple projects.

Can I restructure an existing Copilot agent to meet repository conventions?

Yes, you can restructure an existing Copilot agent to meet repository conventions by applying a structured scaffolding process. This involves updating the YAML frontmatter, refining the prompt body, and ensuring the file is correctly saved under the .github/agents path.

What are the steps to build and validate a Copilot agent prompt?

Building a Copilot agent prompt involves a step-by-step workflow: clarify intent, choose a filename, write YAML frontmatter, craft the prompt body, and validate the configuration. The completed profile is then saved under .github/agents with a descriptive description.