skills.rest
marktantongco
marktantongcoCOMMUNITY
Explore all skills by this author →
GitHub

skill-creator

Create and improve AI skill definitions with evaluation workflows.

2|Updated Mar 31, 2026
Search Tags:#automation#benchmarking#evaluation#packaging#skill-creation#prompt-engineering
One-click install
npx skills add https://github.com/marktantongco/promptc-os --skill skill-creator-marktantongco
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/marktantongco/promptc-os/tree/main/skills/skill-creator
Command: npx skills add https://github.com/marktantongco/promptc-os --skill skill-creator-marktantongco

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Reduces the friction of authoring, testing, and improving AI Skill units by guiding intent capture, drafting SKILL.md content, creating reproducible test cases, running paired with-skill and baseline evaluations, and aggregating results so teams can reliably measure and iterate on skill quality.

Core Features & Use Cases

  • Skill authoring & drafting: Interview-style intent capture, structured SKILL.md templates, and guidance for trigger descriptions and expected output formats.
  • Evaluation orchestration & benchmarking: Automates spawning with-skill and baseline runs, captures timing/tokens, grades outputs, aggregates pass rates, and surfaces deltas.
  • Improvement tooling & packaging: Includes scripts to optimize descriptions, generate review viewers, run blind comparisons, analyze results, and package tested skills for distribution.
  • Use Case: Turn a user's description of a workflow into a validated skill by drafting the skill, creating test prompts, running parallel evaluations, reviewing results in the viewer, and iterating until the skill meets acceptance criteria.

Quick Start

Describe the skill intent, triggering contexts, expected output format, and provide 2–3 example prompts so the assistant can draft SKILL.md, create eval cases, and start the evaluate→improve 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 test AI skills with reproducible evaluation workflows?▼

You can create and test AI skills by capturing intent through an interview-style process, drafting SKILL.md content, generating reproducible test cases, and running paired with-skill and baseline evaluations to measure deltas.

How does benchmarking work for AI prompt engineering and skill optimization?▼

Benchmarking for prompt engineering works by automating parallel with-skill and baseline runs, capturing timing and token usage, grading outputs, and aggregating pass rates to surface measurable deltas for skill optimization.

What's the best way to package tested AI skills for local CLI distribution?▼

The best way to package tested AI skills is to run evaluation loops until acceptance criteria are met, then use built-in packaging scripts with local CLI integrations to prepare the validated skill definitions for distribution.

Do I need YAML to author and iterate on production-grade AI skills?▼

Yes, you need YAML dependencies to author and iterate on production-grade AI skills, as the skill creation and evaluation workflows rely on YAML for configuration, test case definitions, and packaging automation.

Can I automate blind comparisons and review viewers when evaluating AI skills?▼

Yes, you can automate blind comparisons and generate review viewers during AI skill evaluation, using included scripts to analyze results, optimize descriptions, and visually inspect graded outputs for iterative improvements.

Why are my AI skill descriptions not triggering correctly in production scenarios?▼

AI skill descriptions may not trigger correctly if they lack structured trigger contexts or optimized expected output formats, which can be resolved by using description optimization scripts and reviewer-driven iterations.

Related Skills

Explore More Software Engineering Skills→
saajunaidsaajunaid

skill-creator

Draft, evaluate, and refine Claude Skills with SKILL.md and eval cycles.

Community
Advanced
gesmith0606gesmith0606

skill-creator

Create, evaluate, and iteratively improve SKILL.md descriptions with Python scripts.

Community
Advanced
dparedesidparedesi
1

skill-builder

Create, evaluate, and improve Claude Skills using templates and scoring rubrics.

Community
Advanced
coreyjacoreyja
12

skill-writer

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

Community
Basic