raphaelmansuy/machine-learning-feature-selection · Archived

skill-creator

Use this skill to create new Agent Skills for GitHub Copilot. It guides you through the process of setting up the directory structure and the SKILL.md file.

First seen May 26, 2026

Installation

$ npx skills add raphaelmansuy/machine-learning-feature-selection --skill skill-creator

Stronger alternatives

This repository is archived — consider an actively maintained alternative.

Similar popular skills

Related neighbors and high-traction skills in the same topics — useful to compare before installing.

More details

Agent compatibility

Declared targets from SKILL.md / docs. Unmarked agents are not listed — the skill may still install via the CLI.

Claude Code Not declared
Cursor Not declared
Codex Not declared
GitHub Copilot Declared
Windsurf Not declared
Gemini CLI Not declared
Cline Not declared
OpenCode Not declared

Repository health

License LICENSE
Default branch main
Open issues 0
Status Archived

Skill metadata

Parsed from SKILL.md frontmatter.

Declared agents github-copilot

Package contents

Files included with this skill beyond the listing page.

  • skill md SKILL.md 2,707 B
  • docs SUMMARY.md 177 B

History

  1. First seen on skills.sh
  2. First recorded snapshot · 2 installs

SKILL.md

Skill Creator

This skill helps you create new Agent Skills for this repository. Agent Skills are specialized, repeatable tasks that GitHub Copilot can handle in agent mode.

Why Agent Skills?

  • Specialized Workflows: Handle tasks like debugging or testing without repetitive prompting.
  • Open Standard: Uses the Anthropic standard for portability across tools (VS Code, Copilot CLI, Cursor).
  • Domain-Specific: Injects repo-specific instructions and resources on demand.
  • Context Efficiency: Reduces context window bloat by loading only relevant skills.

Mental Model

Agent Skills are repository-level extensions.

  • A skill is a directory under .github/skills/.
  • Each skill contains a mandatory SKILL.md with YAML frontmatter (name, description) and a Markdown body.
  • Copilot matches your request to a skill description and injects the content into its context.

Steps to Create a New Skill

  1. Choose a Name: Use lowercase and hyphens (e.g., my-new-skill).
  2. Create Directory: Create a directory at .github/skills/<skill-name>/.
  3. Create SKILL.md: Create a SKILL.md file inside that directory.
  4. Add Metadata: Include YAML frontmatter with name and description.

- name: The name of the skill (same as the directory name). - description: A clear, prompt-aligned description (e.g., "Use for debugging failing GitHub Actions").

  1. Add Instructions: Write the markdown body with clear steps, examples, and guidance.
  2. Optional Resources: Add scripts, data files, or tools in the same directory.

Survival Kit

  • Day 0: Install VS Code Insiders and enable Copilot.
  • Week 1: Create your first skill and confirm it autoloads by writing a matching prompt.
  • Week 2: Build production skills with scripts and share them in the team repo.

Best Practices

  • One Job Per Skill: Keep skills focused. Split them if they become too large.
  • Clear Descriptions: Vague descriptions cause irrelevant loading.
  • No Secrets: Never embed secrets; use environment variables.
  • Compactness: Keep skills compact to save context.

Example SKILL.md Template

---
name: <skill-name>
description: <clear-description-of-the-skill>
---
<instructions-and-examples-go-here>

Debugging & Observability

  • Enable Copilot debug mode: copilot.chat.debug=true.
  • Check the VS Code output panel logs for skill injection.
  • Use /explain to see how the skill is being applied.