skills.rest
yashs33244
yashs33244COMMUNITY
Explore all skills by this author →
GitHub

skill-creator

Draft SKILL.md files and run trigger evaluations for Claude skills.

1|Updated May 5, 2026
Search Tags:#benchmarking#claude skills#skill authoring#evals#trigger optimization#description tuning#iteration loop
One-click install
npx skills add https://github.com/yashs33244/my-mac-claude --skill skill-creator-yashs33244
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/yashs33244/my-mac-claude/tree/main/skills/anthropic/skill-creator
Command: npx skills add https://github.com/yashs33244/my-mac-claude --skill skill-creator-yashs33244

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Skill-creator helps you turn an idea for a Claude skill into a usable, reliable skill by guiding you through drafting, creating test prompts, running trigger evaluations, and iterating based on results.

Core Features & Use Cases

  • End-to-end skill drafting: Convert a user’s desired behavior into a complete SKILL.md with a structured workflow.
  • Evaluation-driven iteration: Generate quantitative trigger eval queries, run Claude to test triggering behavior, and review outputs to measure progress.
  • Benchmarking for performance: Compare with/without-skill behavior using pass rates, timing, and token usage, then analyze patterns.
  • Description optimization: Improve the SKILL.md description to increase correct triggering while reducing false triggers.
  • Process guidance: Helps you adapt the workflow whether you’re creating a brand-new skill, improving an existing one, or skipping evaluation for “just vibe” prototyping.

Quick Start

Use skill-creator when you want to create a new Claude skill or update an existing one by drafting SKILL.md, generating realistic eval prompts, running trigger/eval loops, and iterating until the skill triggers correctly.

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 a new Claude skill from scratch?▼

You can optimize Claude skill triggering by tuning the SKILL.md description, running quantitative trigger eval queries, and reviewing outputs. This process reduces false triggers and increases correct activation rates for your skill.

How does benchmarking work when evaluating a skill?▼

Benchmarking for skill evaluation compares pass rates, timing, and token usage with and without the skill applied. This variance-aware analysis helps you measure performance patterns and quantify skill improvements accurately.

Can I improve an existing skill without starting the evaluation loop from zero?▼

Yes, you can improve an existing skill by updating its SKILL.md and running targeted trigger evaluations. Alternatively, you can skip evaluation entirely for rapid 'just vibe' prototyping if immediate quantitative benchmarking is not required.

What is a trigger eval query and when do I need it for skill authoring?▼

A trigger eval query is a test prompt used to verify if a skill activates correctly under specific conditions. You need it during skill authoring to measure triggering accuracy, review outputs, and ensure measurable progress before deployment.

Related Skills

Explore More Software Engineering Skills→
javierhbrjavierhbr

platform-contextualizer

Document existing platform state and draft a baseline for Spec-Driven Development.

Community
Advanced
coreyjacoreyja
12

skill-writer

Draft SKILL.md frontmatter and structure guidance for Claude Code skills.

Community
Basic
tnunamaktnunamak

skill-creator

Convert user intent into a trigger-optimized SKILL.md with evaluation workflows.

Community
Advanced
felipesantiago-coderfelipesantiago-coder

skill-creator

Create AI skills with structured evaluation and benchmarking workflows for GLM Code.

Community
Advanced