agent-conventions

Generate standardized .agent.md agent templates with required frontmatter for GitHub Copilot Chat.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Inconsistent agent definitions, unclear delegation patterns, and missing structural conventions make agents fragile and hard to discover or maintain; this Skill provides a single source of truth for how to define, name, and structure agents and subagents for GitHub Copilot Chat in VS Code.

Core Features & Use Cases

  • Provides a user-facing agent template and a subagent template with required frontmatter fields and recommended tools to ensure consistent discovery and behavior.
  • Defines architecture guidance for lean orchestrators and specialized subagents, including delegation patterns, handoffs, and the three-layer test for uniqueness.
  • Documents file naming conventions (PascalCase .agent.md), required frontmatter keys (name, description, tools, agents, user-invokable), and guidance on constraints, skills references, and tool lists for reliable orchestration.

Quick Start

Ask the assistant to generate a new PascalCase .agent.md file using the provided user-facing agent template and include explicit tools, constraints, and skills references.

Frequently Asked Questions about agent-conventions

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

FAQPage Schema
What are the required frontmatter attributes for defining GitHub Copilot Chat agents in VS Code?

Defining GitHub Copilot Chat agents requires explicit frontmatter attributes including name, description, tools arrays, user-invokable flags, and clear constraints. Using PascalCase filenames for .agent.md files ensures predictable orchestration and reliable agent discovery within VS Code.

How do I structure a lean orchestrator and subagent architecture for AI agents?

Structuring lean orchestrator and subagent architectures requires defining specialized subagents with explicit delegation patterns, handoffs, and tool access constraints. Applying the three-layer test for uniqueness ensures each subagent has a distinct purpose, preventing overlapping behaviors and fragile orchestration.

When do I need standardized agent templates for GitHub Copilot Chat?

You need standardized agent templates when inconsistent agent definitions, unclear delegation patterns, or missing structural conventions make agents fragile. Applying a single source of truth for naming and structuring agents ensures predictable orchestration and easier maintenance for GitHub Copilot Chat.

Does GitHub Copilot Chat in VS Code support user-invokable flags for subagent delegation?

Yes, GitHub Copilot Chat supports user-invokable flags within .agent.md frontmatter to control subagent delegation. Defining these flags alongside tools arrays and constraints ensures that orchestrators hand off tasks predictably and only invoke subagents when explicitly requested.

What's the best way to decide tool access and constraints for AI subagents?

Deciding tool access and constraints for AI subagents requires evaluating the subagent's specialized role and applying the three-layer uniqueness test. Explicitly defining allowed tools arrays and clear constraints in the .agent.md frontmatter prevents unauthorized actions and ensures safe orchestration.