skill-creator

Create, refine, and evaluate Claude skills with SKILL.md frontmatter.

2.6k|194|Updated Sep 12, 2025
One-click install
npx skills add https://github.com/heyitsnoah/claudesidian --skill skill-creator-heyitsnoah
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/heyitsnoah/claudesidian/tree/main/.agents/skills/skill-creator
Command: npx skills add https://github.com/heyitsnoah/claudesidian --skill skill-creator-heyitsnoah

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Creating and iterating high-quality Claude skills is complex and time-consuming. This Skill provides a repeatable workflow to draft, test, and improve skills, with guidance on how to structure SKILL.md and use evaluation tooling.

Core Features & Use Cases

  • Supports end-to-end skill creation, iterative testing, and structured evaluation workflows.
  • Provides guidance for running evals, benchmarking variation, and optimizing triggering accuracy.
  • Includes templates and tooling references to streamline skill improvements.

Quick Start

Draft a skill concept, write a first SKILL.md, run evals, review results, and iterate until triggering and quality meet your criteria.

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 Claude skills end-to-end?

To create and test Claude skills, draft a SKILL.md file with valid frontmatter, add optional bundles to scripts, references, and assets, then run evaluations and benchmark performance to iterate on quality and triggering accuracy.

What is a SKILL.md file and when do I need it for prompt engineering?

A SKILL.md file is the core format for defining Claude skills, requiring a name and description in its frontmatter. You need it when creating, refining, or evaluating automated skill workflows to structure triggering logic and bundle dependencies.

What's the best way to optimize skill triggering descriptions and benchmark performance?

The best way to optimize skill triggering descriptions is to run structured evaluations using evals.json artifacts, review timing and benchmark data, then iteratively refine the SKILL.md frontmatter description until triggering accuracy meets your criteria.

Do I need pyyaml to run skill evaluation and benchmarking workflows?

Yes, pyyaml is a required dependency for this skill creation workflow. You need it installed in your environment to parse SKILL.md frontmatter and process evaluation artifacts like evals.json during benchmarking and optimization.

Can I use this workflow to modify existing skills instead of drafting new ones?

Yes, this workflow supports modifying existing skills. You can load an existing SKILL.md, adjust its frontmatter or bundled components in scripts, references, and assets, then run evals to benchmark changes and validate improvements before finalizing.

Why are my skill evaluations not triggering correctly during benchmark testing?

Skill evaluations may not trigger correctly if the SKILL.md frontmatter description lacks specific keywords or is too generic. Running structured evals and reviewing evals.json benchmark data helps identify triggering gaps so you can refine the description for accuracy.