skills.rest
nopperabbo
nopperabboCOMMUNITY
Explore all skills by this author →
GitHub

skill-creator

Guides end-to-end skill creation with drafting, testing, evaluation, and automated description improvement workflows.

7|3|Updated May 4, 2026
Search Tags:#automation#workflow#evaluation#skill-creation#prompt-engineering
One-click install
npx skills add https://github.com/nopperabbo/codebuddy2api --skill skill-creator-nopperabbo
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/nopperabbo/codebuddy2api/tree/main/opencode-config/skills/skill-creator
Command: npx skills add https://github.com/nopperabbo/codebuddy2api --skill skill-creator-nopperabbo

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Efficiently create, iterate on, and evaluate Claude-style skills. This Skill Unit provides a guided workflow to draft new skills, test them with evaluative prompts, compare performance, and progressively improve their descriptions and triggering accuracy.

Core Features & Use Cases

  • Launch a full skill creation cycle from draft to evaluation to optimization.
  • Run controlled evals, analyze results, and iterate on prompts, test prompts, and descriptions.
  • Bundle scripts, references, and assets for repeatable skill deployment and scaling.

Quick Start

Write a draft skill, set up an eval set, and run the evaluation-then-improve loop to refine the skill description.

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 Claude skills from scratch?▼

To create and test Claude skills, you draft a skill, set up an evaluation set, and run an evaluation-then-improve loop to refine prompts and descriptions. This guided workflow facilitates end-to-end skill creation through interview, draft, test, and optimize cycles.

What is the best way to evaluate prompt engineering workflows?▼

The best way to evaluate prompt engineering workflows is by running controlled evals, analyzing aggregated results, and iterating on test prompts. This approach enables structured evaluation prompts and progressive improvement of triggering accuracy.

Do I need PyYAML to automate skill creation and evaluation?▼

Yes, you need PyYAML installed to automate skill creation and evaluation. The skill-creator scripts rely on this dependency for test generation, evaluation execution, reporting, and automated description improvements.

Can I bundle scripts and references for repeatable skill deployment?▼

Yes, you can bundle scripts, references, and assets for repeatable skill deployment and scaling. This allows you to package the structured components needed to consistently trigger and execute the skill across different environments.

How does automated description improvement work for Claude skills?▼

Automated description improvement works by integrating with skill-creator scripts to analyze evaluation results and progressively refine the skill description. This loop enhances triggering accuracy based on aggregated test performance data.

Related Skills

Explore More Software Engineering Skills→
samjhechtsamjhecht
4

initialize-governance

Generate Constitution, Roadmap, and process documentation templates for project governance.

Community
Advanced
LucasMatuszewskiLucasMatuszewski
2

skill-creator

Design, draft, and iteratively improve Claude skills with SKILL.md frontmatter.

Community
Advanced
SJTU-IPADSSJTU-IPADS
7

senior-prompt-engineer

Design, analyze, and optimize prompts with evaluation frameworks and structured outputs.

Official
Advanced
arzzzaearzzzae
1

skill-creator

Draft, test, and refine Copilot skills through an end-to-end evaluation loop.

Community
Advanced