skill-creator

Name: skill-creator
Availability: InStock
Author: yurinakayama

Create, evaluate, and package Claude skills into .skill archives.

Updated Jan 2, 2026

Search Tags:#automation #workflow #evaluation #benchmark #skill #skill-creator

Ready to run

npx skills add https://github.com/YuriNakayama/research --skill skill-creator-yurinakayama

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/YuriNakayama/research/tree/main/.claude/skills/skill-creator
Command: npx skills add https://github.com/YuriNakayama/research --skill skill-creator-yurinakayama

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pyyaml.

What problem does it solve?

This Skill provides a structured framework and tooling to create, evaluate, and iteratively improve Claude skills from scratch, enabling rapid prototyping, reliable evaluation, and repeatable packaging.

Core Features & Use Cases

Guided workflow for capturing intent, drafting SKILL.md, and wiring up frontmatter named fields.
End-to-end evaluation pipeline: run trigger evaluation, generate reports, and compare against baselines to drive improvements.
Packaging and validation utilities: sanity checks, report generation, and artifact packaging into a distributable .skill file.
Use cases include creating a new skill from scratch, iterating an existing skill, optimizing triggering descriptions, and producing ready-to-share skill artifacts.

Quick Start

Define your idea, write SKILL.md, run validation and evaluation, and package the result into a .skill artifact.

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 package Claude skills from scratch?▼

Creating and packaging Claude skills involves defining your intent, drafting SKILL.md, wiring up frontmatter fields, running validation checks, and packaging the validated output into a distributable .skill archive.

What is the best way to evaluate and benchmark a Claude skill?▼

The best way to evaluate a Claude skill is to run the end-to-end evaluation pipeline, which executes trigger evaluations, generates reports, and compares results against baselines to drive iterative improvements.

How does trigger evaluation work when optimizing skill discovery?▼

Trigger evaluation works by running defined test prompts against your skill to generate reports and compare baseline metrics, allowing you to optimize the triggering description for better discovery and reliable activation.

Do I need pyyaml to automate skill creation and validation workflows?▼

Yes, pyyaml is required to automate skill creation and validation workflows, as it handles YAML parsing for frontmatter fields and configuration files during the automated validation and packaging workflow.

Can I iterate on an existing skill and repackage it into a .skill file?▼

Yes, you can iterate on an existing skill by modifying the SKILL.md file, re-running trigger evaluations to compare against baselines, and repackaging the validated updates into a new .skill archive.