skills.rest
lucasgaldinos
lucasgaldinosCOMMUNITY
Explore all skills by this author →
GitHub

skill-creator

Automate creating, testing, and refining Claude skills with SKILL.md and evals.

Updated Apr 2, 2026
Search Tags:#automation#workflow#testing#Claude#evaluation#benchmark#skill
One-click install
npx skills add https://github.com/lucasgaldinos/my-skills --skill skill-creator-lucasgaldinos
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/lucasgaldinos/my-skills/tree/main/skills/skill-creator
Command: npx skills add https://github.com/lucasgaldinos/my-skills --skill skill-creator-lucasgaldinos

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

The Skill Creator streamlines the end-to-end process of creating, testing, and iterating Claude skills. It guides you from drafting SKILL.md to running evaluations, collecting results, and refining the description and execution flow based on feedback.

Core Features & Use Cases

  • Draft and organize SKILL.md with optional resources in scripts/, references/, and assets/ directories.
  • Run trigger-evaluation loops to assess when a skill should trigger and how well it performs.
  • Generate, review, and act on benchmark reports to compare configurations and iterations.
  • Iterate rapidly by improving the skill description, auditing results, and bundling useful scripts for reuse.

Quick Start

Draft a new skill using SKILL.md, then run the evaluation loop to test and iterate until you’re satisfied.

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?▼

You can automate Claude skill creation by drafting a SKILL.md file, running trigger evaluations to test performance, and iterating on descriptions using the run_eval and analyzer tooling loop.

What is the best way to test and iterate on Claude skills?▼

The best way to test Claude skills is running trigger-evaluation loops to assess trigger accuracy, then refining the execution flow and description based on benchmark reports and collected feedback.

How do I generate benchmarks comparing multiple Claude skill configurations?▼

You generate benchmarks for multiple skills by running evaluation loops across configurations, which produces reports you can review to compare iterations and trigger performance metrics.

Do I need YAML to validate SKILL.md frontmatter?▼

Yes, YAML is required to validate frontmatter. Your SKILL.md must include valid name and description fields, while scripts, references, and assets directories remain optional dependencies.

What components are required when building a Claude skill?▼

Building a Claude skill requires a SKILL.md file with validated name and description frontmatter. Optional components include scripts, references, and assets directories to organize supporting resources.

Why does my Claude skill trigger evaluation loop fail?▼

Trigger evaluation loops fail when SKILL.md frontmatter lacks required name or description fields. Validate your YAML frontmatter and ensure all optional scripts and assets are correctly bundled before running evals.

Related Skills

Explore More Software Engineering Skills→
srprasannasrprasanna
11

minimal-example

Create a minimal Claude Skill skeleton with required frontmatter fields.

Community
Basic
j1nggj1ngg
55

autoresearch

Benchmark and improve existing repo skills through repeated evals and score tracking.

Community
Advanced
Power-AgentPower-Agent
58

skill-creator

Create, evaluate, and optimize Claude skills through a repeatable workflow.

Official
Intermediate
dariopalminiodariopalminio
1

skill-creator

Orchestrate Claude skill creation, evaluation, and packaging into .skill files.

Community
Advanced