skills.rest
mammut001
mammut001COMMUNITY
Explore all skills by this author →
GitHub

skill-creator

Draft SKILL.md files, run evaluations, and iterate on Claude skills.

2|Updated Mar 13, 2026
Search Tags:#workflow#evaluation#iteration#benchmark#skill-creation#prompt-engineering
One-click install
npx skills add https://github.com/mammut001/pipi-shrimp-agent --skill skill-creator-mammut001
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/mammut001/pipi-shrimp-agent/tree/main/src-tauri/skills/skill-creator
Command: npx skills add https://github.com/mammut001/pipi-shrimp-agent --skill skill-creator-mammut001

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires anthropic, yaml.

What problem does it solve?

The Skill Creator enables teams to quickly craft, test, and refine Claude skills through an end-to-end, repeatable workflow that couples drafting with rigorous evaluation and benchmarking.

Core Features & Use Cases

  • Create new skills by drafting SKILL.md with frontmatter and instructions.
  • Run structured evals to measure trigger accuracy and performance.
  • Iterate descriptions, tests, and references to improve discovery and effectiveness.
  • Use eval-viewer and benchmark tooling to review results and compare iterations.

Quick Start

Write a new skill’s SKILL.md with frontmatter, draft the body, run an initial eval, and begin the iterative improvement loop.

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 Claude skills systematically?▼

You create and evaluate Claude skills by drafting a SKILL.md file with frontmatter, running structured evals to measure trigger accuracy, and iterating based on benchmark results to optimize performance.

What is a structured workflow for prompt engineering iteration?▼

A structured prompt engineering iteration workflow couples drafting instructions with rigorous evaluation and benchmarking, allowing you to quantify improvements and maintain auditable history for continuous optimization.

How do I benchmark skill performance across multiple iterations?▼

To benchmark skill performance across iterations, use the built-in eval-viewer and benchmark tooling to review results, compare changes, and maintain an auditable history of trigger accuracy and effectiveness.

Does the Skill Creator workflow require YAML and Anthropic dependencies?▼

Yes, the Skill Creator workflow requires Anthropic and YAML dependencies to orchestrate the end-to-end lifecycle of building, evaluating, and improving Claude skills.

Can I test trigger descriptions to improve skill discovery?▼

Yes, you can test trigger descriptions to improve skill discovery by running structured evals that measure trigger accuracy and iterating the descriptions based on the evaluation feedback.

Related Skills

Explore More Software Engineering Skills→
avibebuilderavibebuilder
113

skill-creator

Create testable Claude Code SKILL.md files with frontmatter and evaluation benchmarks.

Official
Advanced
qiushibangqiushibang

skill-creator

Create, modify, and improve AI agent skills through eval-driven workflows.

Community
Advanced
Let-s-internLet-s-intern
8

skill-creator

Create, test, and refine Claude skills into SKILL.md files with automated evals.

Official
Advanced
zhbonerzhboner

skill-creator

Draft SKILL.md files and refine them through prompt-based test evaluations.

Community
Advanced