ghc-prompt-files

Guides learners through building reusable GitHub Copilot prompt files with YAML frontmatter and input parameters.

2|Updated Jun 1, 2026
One-click install
npx skills add https://github.com/jay-steenbergen/MSSAMentorAgent --skill ghc-prompt-files-jay-steenbergen
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ghc-prompt-files
Source: https://github.com/jay-steenbergen/MSSAMentorAgent/tree/main/.github/skills/tracks/github-copilot/ghc-prompt-files
Command: npx skills add https://github.com/jay-steenbergen/MSSAMentorAgent --skill ghc-prompt-files-jay-steenbergen

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Engineers repeatedly type the same multi-step requests into Copilot Chat, wasting effort and producing inconsistent results. This Skill teaches learners to templatize recurring tasks as version-controlled .prompt.md files with parameters, invoked on demand from Chat. ## Core Features & Use Cases - Prompt File Authoring: Create .github/prompts/*.prompt.md files with YAML frontmatter (mode, description, model) and ${input:...} parameter placeholders. - Mode Selection: Learn when to use agent mode (multi-file orchestration), edit mode (in-place file changes), or ask mode (chat-only). - Surface Decision Tree: Build a decision framework distinguishing prompt files vs slash commands vs instruction files vs custom agents. - Use Case: A developer who frequently scaffolds pytest fixtures, writes docstrings, or adds CLI commands builds three prompt files, invokes them via /name in Chat, and codifies which Copilot surface fits each kind of task. ## Quick Start Ask the Mentor to start the ghc-prompt-files project and help you create your first reusable .prompt.md file with input parameters.

Frequently Asked Questions about ghc-prompt-files

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

FAQPage Schema
How do I create a reusable prompt file for GitHub Copilot?

Create a file at .github/prompts/<name>.prompt.md with YAML frontmatter specifying mode and description, then write the prompt body with ${input:variableName} placeholders. Invoke it from Copilot Chat by typing /<name> and providing the parameter values when prompted.

What is the difference between a prompt file and a slash command in Copilot?

Slash commands like /explain or /fix are built-in and fixed. Prompt files are user-defined, version-controlled with the repo, and accept parameters through ${input:...} placeholders. The invocation shape is the same, but ownership and parameterization differ.

When should I use agent mode vs edit mode in a prompt file?

Use agent mode when the task requires tools like reading files, writing multiple files, or running commands, such as scaffolding a CLI command with tests. Use edit mode when the result should modify the current file directly, like adding a docstring to a selected function.

Why does my prompt file not appear in the Copilot Chat picker?

VS Code caches the prompt list, so a newly created .prompt.md file may not appear immediately. Restart the Chat session to refresh the picker, and confirm the file lives in .github/prompts/ with the .prompt.md extension.

When should I use a custom agent instead of a prompt file?

Use a custom agent when the task needs a persistent persona across a multi-turn conversation, such as an API design reviewer. Prompt files fit one-shot templated tasks; agents fit ongoing roles with sustained context.