skills.rest
otey247
otey247COMMUNITY
Explore all skills by this author →
GitHub

skill-creator

Build, test, and optimize Copilot agent skills with SKILL.md workflows.

1|Updated Mar 30, 2026
Search Tags:#workflow#evaluation#benchmark#skill#agent-skills
One-click install
npx skills add https://github.com/otey247/agent-skill-creator --skill skill-creator-otey247
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/otey247/agent-skill-creator/tree/main/.github/skills/skill-creator
Command: npx skills add https://github.com/otey247/agent-skill-creator --skill skill-creator-otey247

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pyyaml, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

The Skill Creator provides a complete workflow for authoring, testing, benchmarking, and improving Copilot agent skills and their descriptions.

Core Features & Use Cases

  • Create new skills from scratch and iteratively refine them.
  • Run evaluations to validate skill behavior, benchmark performance, and optimize triggering accuracy.
  • Use the built-in workflows to track iterations, compare versions, and generate action-oriented improvements to SKILL.md.

Quick Start

Use this skill to define, test, and optimize Copilot agent skills. Start by creating a SKILL.md with a clear name and description, then run the included eval and benchmark scripts to iterate.

Frequently Asked Questions about skill-creator

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

FAQPage Schema
How do I build and test Copilot agent skills end-to-end?▼

To build Copilot agent skills end-to-end, you define a SKILL.md file with a clear name and description, then run the included evaluation and benchmark scripts to validate behavior and optimize triggering accuracy.

What is the best way to benchmark agent skills and optimize triggering rules?▼

Benchmarking agent skills involves running evaluation protocols to validate behavior, compare version iterations, and generate action-oriented improvements to SKILL.md, ensuring robust production-grade triggering accuracy.

Do I need pyyaml to create and evaluate Copilot agent skills?▼

Yes, pyyaml is a required dependency to run the workflows, evaluation scripts, and benchmarking metrics for authoring and optimizing Copilot agent skills.

Can I use this workflow to iteratively refine agent skills for production?▼

Yes, you can iteratively refine agent skills for production by tracking iterations, comparing versions, and applying generated action-oriented improvements to the SKILL.md descriptions.

What are the limitations when benchmarking agent skills with this workflow?▼

The workflow focuses specifically on authoring and benchmarking Copilot agent skills using SKILL.md files and pyyaml, targeting software engineers and product managers rather than general unrelated automation tasks.

Does the Skill Creator work for product managers designing agent capabilities?▼

Yes, the Skill Creator is designed for practical use by both software engineers and product managers to design, test, and optimize robust Copilot agent capabilities across workflows.

Related Skills

Explore More Software Engineering Skills→
TerminalGambitTerminalGambit

Presentation Framework

Generate branded HTML slide decks from YAML and JSON metrics.

Community
Advanced
Division6066Division6066

skill-creator

Guide authors through SKILL.md creation, testing, and evaluation cycles.

Community
Advanced
nanorepublicananorepublica
61

Global Tech Stack

Validate code against centralized global tech stack standards.

Community
Basic
pagopapagopa
1

TechAISkillCreator

Create, refine, and benchmark AI skills with automated testing.

Official
Advanced