skill-creator

Coordinate end-to-end creation, testing, and iteration of Codex Skills.

1|Updated Apr 10, 2026
One-click install
npx skills add https://github.com/EzioPeter/codex_skills --skill skill-creator-eziopeter
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/EzioPeter/codex_skills/tree/main/skill-creator
Command: npx skills add https://github.com/EzioPeter/codex_skills --skill skill-creator-eziopeter

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill helps teams create, test, and iteratively improve Codex Skills by codifying the process into SKILL.md, eval prompts, and evaluation workflows.

Core Features & Use Cases

  • Draft SKILL.md content and generate initial prompts.
  • Create test prompts, run evals, and collect quantitative and qualitative feedback.
  • Iterate on skill description, templates, and tooling based on evaluation results.
  • Package finalized skills for distribution and reuse.

Quick Start

Draft a new skill with SKILL.md, run a few eval prompts, and begin the iterative improvement loop.

Frequently Asked Questions about skill-creator

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

FAQPage Schema
How do I create and test Codex Skills end-to-end?

To create and test Codex Skills, draft a SKILL.md file, generate initial prompts, run controlled eval tests, and aggregate results to iteratively refine descriptions and templates before packaging for distribution.

What is the best way to iterate on prompt engineering for skill creation?

The best way to iterate on prompt engineering is to design specific eval prompts, run controlled tests to collect quantitative and qualitative feedback, and use those aggregated results to drive improvements to your templates.

Do I need pyyaml to package and distribute Codex Skills?

Yes, pyyaml is required as a dependency to coordinate the skill creation workflow, parse configuration formats, and package finalized skills with clear validation and benchmarking for distribution.

Can I run controlled evals to improve an existing skill description?

Yes, you can run controlled evals against existing skills. The system collects quantitative and qualitative feedback from test prompts to drive targeted improvements in your skill description and tooling.

What is a SKILL.md file used for in skill creation?

A SKILL.md file is used to codify the skill creation process, define core features, and document the finalized skill content ready for validation, benchmarking, and distribution.

Are there limitations to automating skill creation with a training loop?

The training loop relies on the quality of your designed eval prompts; poorly constructed tests yield inaccurate feedback, limiting the effectiveness of iterative improvements to your skill templates and descriptions.