skill-creator

Automate Claude skill creation, evaluation, and iterative improvement from SKILL.md.

Updated May 12, 2026
One-click install
npx skills add https://github.com/YeomanYe/skills --skill skill-creator-yeomanye
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/YeomanYe/skills/tree/main/skill-creator
Command: npx skills add https://github.com/YeomanYe/skills --skill skill-creator-yeomanye

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Creating, validating, and improving Claude skills from a shared SKILL.md is manual, error-prone, and slow. This skill automates the entire lifecycle, enabling rapid iteration from draft to production-ready units.

Core Features & Use Cases

  • Automated SKILL.md scaffolding: generate and validate frontmatter and structure.
  • Integrated eval loop: run prompts, collect expectations, and produce grading data.
  • Description optimization: propose concise, trigger-friendly SKILL descriptions based on results.
  • Bundled resources: reuse deterministic scripts and references for consistent tasks.
  • Use case: a product team creates a new skill, automates evaluation, and iterates until performance meets quality gates.

Quick Start

Run the skill-creation workflow to scaffold a new skill, run evaluations, and iterate on the description.

Frequently Asked Questions about skill-creator

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

FAQPage Schema
How do I automate Claude skill creation and evaluation from a SKILL.md file?

Automating Claude skill creation involves running a workflow that scaffolds SKILL.md frontmatter, executes integrated evaluation loops, and iteratively improves descriptions until production-ready. This eliminates manual validation and accelerates team workflows.

What is the best way to validate and improve SKILL.md descriptions for quality assurance?

The best way to validate descriptions is using an integrated evaluation loop that runs prompts, collects expectations, and produces grading data. The system then proposes concise, trigger-friendly SKILL descriptions based on benchmark results.

Do I need YAML frontmatter to scaffold a new skill?

Yes, YAML frontmatter with a name and description is required at SKILL.md to scaffold a new skill. Optional resources like deterministic scripts, references, and assets can be bundled to support consistent task execution.

Can I use automated evaluations to meet quality gates for my product team's skills?

Yes, your product team can use automated evaluations to run benchmarks and collect feedback loops. You can iterate on the skill until its performance consistently meets your defined quality gates.

What are the limitations of using YAML for skill automation?

YAML frontmatter must strictly contain a name and description to function correctly. If the structure is invalid, scaffolding fails, so you must ensure proper formatting before running the evaluation and improvement workflow.

How does the built-in review UI work for skill evaluation?

The built-in review UI displays grading data generated from running prompts and collecting expectations. It allows teams to visually inspect evaluation results and iterate on the skill until it reaches production readiness.