skill-creator

Creates and improves self-contained skills with drafting, eval design, benchmarking, and optimization workflows.

Updated Jul 2, 2025

Search Tags:#skill creation #benchmarking #iteration #skill improvement #evals #trigger optimization

Ready to run

npx skills add https://github.com/junechen7414/SpringBoot --skill skill-creator-junechen7414

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/junechen7414/SpringBoot/tree/main/.bob/skills/skill-creator
Command: npx skills add https://github.com/junechen7414/SpringBoot --skill skill-creator-junechen7414

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill helps you create new skills from scratch, refine existing ones, and measure whether they actually work well in practice.

Core Features & Use Cases

Skill Design: Define what a skill should do, when it should trigger, and what outputs it should produce.
Iterative Improvement: Draft, test, evaluate, and rewrite skills based on real usage and feedback.
Benchmarking and Analysis: Run evals, compare with and without the skill, and inspect timing, token use, and pass rates.
Trigger Optimization: Improve the skill description so it activates more reliably for the right user requests.
Use Case: If you want to turn a repeated workflow into a reusable skill, this guide walks you through the full lifecycle from idea to polished result.

Quick Start

Use the skill creator to help me design, test, and improve a skill for my workflow.

Frequently Asked Questions about skill-creator

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

FAQPage Schema

How do I create a reusable skill from a repeated workflow?▼

You can create a reusable skill by defining its trigger conditions and expected outputs, drafting the initial instructions, and structuring the skill metadata for self-contained execution. This process turns repeated workflows into polished, reliable skills.

What is the best way to test if a skill triggers accurately for the right user requests?▼

Testing trigger accuracy requires trigger optimization, which involves improving the skill description and running evals. This ensures the skill activates reliably only when matching user requests occur.

How do I benchmark and analyze skill performance during iteration?▼

Benchmarking skill performance involves running evals, comparing results with and without the skill, and inspecting timing, token use, and pass rates. This analysis identifies areas for iterative improvement based on real feedback.

Can I use this to refine an existing skill instead of building a new one?▼

Yes, you can refine an existing skill by rewriting instructions, testing trigger accuracy, and iterating based on real usage. It supports editing workflows to measure whether existing skills work well in practice.

What components do I need to design self-contained skills?▼

Designing self-contained skills requires structuring skill metadata, generating evals, and aggregating benchmark data. These components ensure reliable skill development from the initial idea to the polished result.