skills.rest
weidbchina
weidbchinaCOMMUNITY
Explore all skills by this author →
GitHub

skill-creator

Automate AI skill creation, updates, and evaluations with SKILL.md metadata.

Updated Apr 13, 2026
Search Tags:#tooling#evaluation#iteration#creation#skill#prompt-engineering#AI-workflow
One-click install
npx skills add https://github.com/weidbchina/connector --skill skill-creator-weidbchina
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/weidbchina/connector/tree/main/.trae/skills/skill-creator
Command: npx skills add https://github.com/weidbchina/connector --skill skill-creator-weidbchina

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Coordinating the end-to-end lifecycle of AI skill creation, improvement, and evaluation is complex and error-prone without a standardized framework that captures structure, triggers, and metadata.

Core Features & Use Cases

  • Create new skills from scratch with a clearly defined SKILL.md frontmatter and a structured body that guides behavior in context.
  • Modify, extend, and optimize existing skills; run evals, benchmark performance with variance analysis, and refine descriptions for better triggering accuracy.
  • Use bundled tooling such as eval-viewer and reference materials to visualize results, iterate quickly, and maintain consistency across skills.

Quick Start

Create a new skill by writing SKILL.md with name and description, then implement the body and add optional bundles (scripts, references, assets) as needed.

Frequently Asked Questions about skill-creator

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

FAQPage Schema
How do I create an AI skill from scratch using SKILL.md?▼

To create an AI skill from scratch, write a SKILL.md file with a YAML frontmatter defining the name and description, implement the body to guide behavior, and add optional scripts, references, or assets as needed.

How does prompt evaluation and variance analysis work for existing AI skills?▼

Prompt evaluation for existing AI skills works by running standardized evals, benchmarking performance with variance analysis, and refining descriptions to improve triggering accuracy and consistency across iterations.

What is the best way to structure a folder for AI workflow automation skills?▼

The best way to structure an AI workflow skill folder is to include a SKILL.md with YAML frontmatter, and group optional bundles into scripts, references, and assets directories to maintain consistency.

Can I use bundled tooling to visualize evaluation results for prompt engineering?▼

Yes, you can use bundled tooling like the eval-viewer to visualize evaluation results for prompt engineering, allowing you to iterate quickly and maintain consistency across multiple AI skills.

Does modifying AI skills require a standardized metadata format?▼

Yes, modifying AI skills requires a standardized metadata format, specifically a structured skill folder with a YAML frontmatter in SKILL.md to drive standardized prompts and manage evaluation results.

Related Skills

Explore More Software Engineering Skills→
MrMoshkovitzMrMoshkovitz

skill-creator

Create Claude skills with YAML frontmatter and optional resource bundles.

Community
Intermediate
coreyjacoreyja
12

skill-writer

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

Community
Basic
rangerrick337rangerrick337
10

skill-creator

Create Claude skills with SKILL.md frontmatter and package them into .skill files.

Community
Advanced
cognitedatacognitedata
5

skill-creator

Create, refine, and benchmark AI skills through iterative AI-guided workflows.

Official
Intermediate