skills.rest
Ahoo-Wang
Ahoo-WangCOMMUNITY
Explore all skills by this author →
GitHub

skill-creator

Create and evaluate Claude skills with SKILL.md and eval workflows.

23|1|Updated Dec 1, 2023
Search Tags:#automation#workflow#benchmarking#evaluation#iteration#skill-creation
One-click install
npx skills add https://github.com/Ahoo-Wang/wow-project-template --skill skill-creator-ahoo-wang
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/Ahoo-Wang/wow-project-template/tree/main/.agents/skills/skill-creator
Command: npx skills add https://github.com/Ahoo-Wang/wow-project-template --skill skill-creator-ahoo-wang

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 helps teams systematically craft Claude skills by guiding them through conceptualizing, drafting, testing, evaluating, and iterating skills from initial idea to refined, production-ready SKILL.md and eval workflows.

Core Features & Use Cases

  • End-to-end skill lifecycle: design, draft, test, evaluate, benchmark, and optimize skills.
  • Iterative improvement: generate data-driven skill updates from evaluation results and history, then re-run the evaluation loop.
  • Bundled resources: scripts, references, and assets that support deterministic tasks, in-depth reading, and output templating.

Quick Start

Start by outlining a skill concept, write a draft SKILL.md, configure evals, and run the evaluation loop to iteratively refine the skill until the results stabilize.

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 evaluate skills from an initial idea?▼

To create and evaluate skills, you outline a concept, draft a SKILL.md file, configure eval sets, and run an evaluation loop to iteratively refine the skill until benchmarking results stabilize.

What is the best way to iteratively improve an existing skill?▼

The best way to iteratively improve an existing skill is to run an evaluation pipeline, generate data-driven updates from evaluation history and variance analysis, then re-run the evaluation loop.

How does benchmarking with variance analysis work for skill evaluation?▼

Benchmarking with variance analysis executes evaluation pipelines across multiple runs, measuring output fluctuations to identify trigger description inconsistencies and guide data-driven skill optimizations.

Do I need to parse YAML frontmatter to build a production-grade skill?▼

Yes, parsing YAML frontmatter is required to configure trigger descriptions and metadata correctly when building and packaging a production-grade SKILL.md file.

Can I bundle scripts and references with my SKILL.md file?▼

Yes, you can bundle scripts, references, and assets with your SKILL.md to support deterministic tasks, provide in-depth reading materials, and template outputs for your workflow.

Related Skills

Explore More Software Engineering Skills→
marcusrbrownmarcusrbrown
23

creating-agent-skills

Create and refine Claude Code Skills with valid YAML frontmatter and XML bodies.

Community
Advanced
I-OnlabsI-Onlabs
3

senior-data-scientist

Design and execute production-grade data science experiments with Python-based tooling.

Community
Advanced
nickstradnickstrad

skill-creator

Generate Claude skill scaffolds with SKILL.md templates and example resources.

Community
Intermediate
openJiuwen-aiopenJiuwen-ai
2.5k

skill-creator

Draft SKILL.md content, generate test prompts, and run trigger-evaluation benchmarks.

Official
Advanced