skills.rest
yusuk2s
yusuk2sCOMMUNITY
Explore all skills by this author →
GitHub

skill-creator

Draft SKILL.md files and benchmark trigger accuracy with automated evals.

Updated Mar 22, 2026
Search Tags:#optimization#evaluation#iteration#benchmark#skill#claude-code
One-click install
npx skills add https://github.com/yusuk2s/knowledge-sync --skill skill-creator-yusuk2s
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/yusuk2s/knowledge-sync/tree/main/.agents/skills/skill-creator
Command: npx skills add https://github.com/yusuk2s/knowledge-sync --skill skill-creator-yusuk2s

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Writing and maintaining effective Claude skills is challenging. This Skill unit provides a structured process to draft SKILL.md, capture intent, and validate triggering through automated evals and benchmarks, enabling rapid, repeatable improvements.

Core Features & Use Cases

  • Capture intent and scope for a new skill, or refine an existing one.
  • Run trigger evaluations to measure how often the skill would be activated.
  • Benchmark performance across configurations (with_skill vs without_skill) and generate actionable improvements.
  • Package skills into distributable formats for sharing and deployment.

Quick Start

Draft your SKILL.md, prepare a small eval set, and run the evaluation loop to start iterative improvements.

Frequently Asked Questions about skill-creator

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

FAQPage Schema
How do I evaluate and improve Claude skill triggering accuracy?▼

Benchmarking Claude skills compares performance across with_skill and without_skill configurations to generate actionable improvements. This method measures triggering accuracy and validates whether the skill produces the intended behavioral changes.

What is the best way to create a structured SKILL.md file for Claude?▼

The best way to create a structured SKILL.md file is to follow a guided process that captures intent and scope before validation. This structured approach ensures your skill definition is clear, repeatable, and ready for automated evaluation.

Do I need YAML to draft and benchmark Claude skills?▼

Yes, YAML is a required dependency to draft and benchmark Claude skills. You need YAML configured in your environment to parse skill metadata and execute the automated evaluation loops effectively.

How do I package Claude skills for sharing and deployment?▼

You package Claude skills for sharing and deployment by using the skill creator's packaging functionality after benchmarking. This process formats your validated SKILL.md and associated scripts into a distributable format for deployment.

Why are my Claude skills not triggering correctly in automated workflows?▼

Claude skills may not trigger correctly due to poorly defined intent or scope in SKILL.md. Running automated trigger evaluations helps identify activation failures and provides actionable data to iterate toward higher triggering accuracy.

Related Skills

Explore More Software Engineering Skills→
xlsamaxlsama

skill-creator

Create and evaluate AI skills with SKILL.md files and baseline tests.

Community
Advanced
gelowzzzgelowzzz

skill-creator

Create, validate, and package Claude Skills into .skill files.

Community
Advanced
rysweetrysweet

model-evaluation-benchmark

Automate end-to-end model evaluation benchmarks and generate reproducible reports.

Community
Advanced
wtgoku-createwtgoku-create

skill-creator

Iteratively refine SKILL.md descriptions through automated trigger evaluations and benchmark outputs.

Community
Advanced