skill-creator

Automate the end-to-end workflow for creating and refining Claude Skills.

Updated Dec 17, 2025

Search Tags:#automation #evaluation #iteration #frontmatter #prompt-engineering #skill-creator

Ready to run

npx skills add https://github.com/Lrmdom/patrimonio-pwa --skill skill-creator-lrmdom

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/Lrmdom/patrimonio-pwa/tree/main/.windsurf/skills/skill-creator
Command: npx skills add https://github.com/Lrmdom/patrimonio-pwa --skill skill-creator-lrmdom

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires anthropic, pyyaml.

What problem does it solve?

Automates the end-to-end workflow for creating, validating, and iterating Claude Skills, helping teams define tests, run evaluations, and improve discoverability.

Core Features & Use Cases

Guided skill creation workflow: draft a new skill, define test prompts, and run evals to validate behavior.
Iterative improvement loop: analyze results, update frontmatter/description, and re-run evaluations to converge on a robust skill.
Trigger optimization and description refinement to improve triggering accuracy and discovery across domains.

Quick Start

Use the skill creator to initiate a new skill from a conversation, then iteratively run evals and improvements until outputs meet acceptance criteria.

Frequently Asked Questions about skill-creator

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

FAQPage Schema

How do I automate prompt engineering and evaluation for Claude Skills?▼

You can automate prompt engineering and evaluation for Claude Skills by running a structured loop that drafts skill prompts, defines test cases, executes evaluations, and iterates on frontmatter to improve triggering accuracy and overall skill performance.

What is the best way to validate and iterate on Claude Skill frontmatter?▼

The best way to validate and iterate on Claude Skill frontmatter is to run an automated evaluation loop that tests prompt behavior, analyzes results, and updates descriptions to ensure robust triggering and discovery across product, engineering, and data domains.

How do I create and test Claude Skills end-to-end from a conversation?▼

You create and test Claude Skills end-to-end by initiating a new skill from a conversation, defining test prompts to validate behavior, and iteratively running evaluations and improvements until the outputs meet your acceptance criteria.

Can I use Anthropic and PyYAML dependencies to optimize skill triggering accuracy?▼

Yes, you can use the Anthropic and PyYAML dependencies to optimize skill triggering accuracy by validating frontmatter structure, running test prompts, and refining skill descriptions through an iterative evaluation loop.

Why does my Claude Skill fail to trigger correctly across different domains?▼

Your Claude Skill may fail to trigger correctly if the frontmatter and descriptions lack optimization, which requires running an iterative evaluation loop to analyze test results, update prompts, and improve discovery across product, engineering, and data domains.