skills.rest
dsteven12
dsteven12COMMUNITY
Explore all skills by this author →
GitHub

skill-creator

Create, test, and improve Cowork skills through automated evals and iteration.

1|Updated Mar 24, 2026
Search Tags:#automation#workflow#evaluation#iteration#benchmark#skill-creation
One-click install
npx skills add https://github.com/dsteven12/airtable-sa-skills --skill skill-creator-dsteven12
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/dsteven12/airtable-sa-skills/tree/main/skills/skill-creator
Command: npx skills add https://github.com/dsteven12/airtable-sa-skills --skill skill-creator-dsteven12

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill provides a structured, repeatable workflow for creating and iterating Cowork skills. It guides you through capturing intent, drafting SKILL.md, running evals, analyzing results, and improving the triggering description to maximize reliability and performance.

Core Features & Use Cases

  • Skill creation pipeline: from initial idea to a testable SKILL.md, with iterative improvement loops.
  • Eval-driven iteration: run trigger evaluations, compare with baselines, and automatically refine prompts and test sets.
  • Benchmarking & reporting: generate quantitative benchmarks and human-readable summaries to track progress over iterations.
  • Description optimization: use built-in helpers to improve trigger descriptions for better accuracy and coverage.
  • Packaging: validate and package skills into distributable .skill files.

Quick Start

Clone this repository, edit the SKILL.md for your own skill, and run the evaluation loop to begin iterating.

Frequently Asked Questions about skill-creator

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

FAQPage Schema
How do I automate AI skill creation and testing workflows?▼

Automating AI skill creation involves capturing intent, drafting a SKILL.md file, running trigger evaluations, and iterating on prompts. This skill provides a structured pipeline to generate testable skills and refine them through automated evals.

What is the best way to optimize AI trigger descriptions for better accuracy?▼

Optimizing trigger descriptions requires running iterative evaluations against a baseline and analyzing the results. Built-in helpers allow you to refine descriptions systematically, maximizing trigger reliability and coverage for your AI skills.

How do I run eval-driven iteration for benchmarking prompt performance?▼

Eval-driven iteration benchmarks prompt performance by running trigger evaluations, comparing outcomes with baselines, and generating quantitative reports. This structured loop tracks progress over iterations to improve overall skill reliability.

Can I use YAML to configure and package distributable skill files?▼

Yes, YAML is required as a dependency to configure skill metadata. The workflow validates your YAML configuration and packages the tested skills into distributable .skill files for deployment.

What is a structured pipeline for improving Cowork skills through automated evaluation?▼

A structured improvement pipeline captures initial intent, drafts the skill definition, runs automated evals, and analyzes benchmark results. This repeatable workflow enables continuous iteration to maximize skill performance and reliability.

Related Skills

Explore More Software Engineering Skills→
yangliu2060yangliu2060
39

youtube-video-analyzer

Extract YouTube transcripts and captions into structured summaries with key points.

Community
Basic
gelowzzzgelowzzz

skill-creator

Create, validate, and package Claude Skills into .skill files.

Community
Advanced
coreyjacoreyja
12

skill-writer

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

Community
Basic
ralphjohn29ralphjohn29
1

skill-creator

Create, test, and benchmark custom AI skills with SKILL.md files.

Community
Advanced