skill-creator

Name: skill-creator
Availability: InStock
Author: Corvalis

Outline a skill workflow from drafting to evaluation with SKILL.md structure.

6|Updated Mar 31, 2026

Search Tags:#automation #workflow #testing #documentation #evaluation #iteration #skill-creation

Ready to run

npx skills add https://github.com/Corvalis-LLC/Crow-Stack --skill skill-creator-corvalis-llc

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/Corvalis-LLC/Crow-Stack/tree/main/skills/skill-creator
Command: npx skills add https://github.com/Corvalis-LLC/Crow-Stack --skill skill-creator-corvalis-llc

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Drafting and refining Claude skills is error-prone and time-consuming. The Skill Creator provides a repeatable workflow to design a SKILL.md, bundle resources, run automated evaluations, and iterate until triggering is reliable.

Core Features & Use Cases

Create new skills from scratch and evolve existing ones through guided prompts and tests.
Run evals to measure triggering accuracy, performance, and robustness.
Benchmark skill descriptions and script usage to identify most effective configurations.
Organize resources in scripts/, references/, and assets/ for on-demand loading.

Quick Start

Describe your intended skill and start the Skill Creator to draft, test with evals, and iterate until triggering is reliable.

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 Claude skill from scratch?▼

Creating a Claude skill involves drafting a SKILL.md, bundling resources in scripts, references, and assets directories, running automated evaluations, and iterating until triggering is reliable.

What's the best way to test skill triggering accuracy?▼

Run automated evaluations to measure triggering accuracy, performance, and robustness, then benchmark skill descriptions and script usage to identify the most effective configurations.

How does automated evaluation work for workflow automation skills?▼

Automated evaluation benchmarks skill descriptions and script usage against specified inputs, outputs, and success criteria to measure triggering accuracy, performance, and robustness.

Do I need pyyaml and anthropic dependencies to build skills?▼

Yes, the Skill Creator requires the anthropic and pyyaml dependencies to function, providing the necessary libraries for skill creation and configuration management.

Can I organize skill resources into scripts, references, and assets directories?▼

Yes, you can organize resources in scripts, references, and assets directories for on-demand loading, enabling efficient bundling and access of optional skill resources.

Why does my skill description not trigger reliably during testing?▼

Skill triggering may be unreliable if the SKILL.md structure and descriptions are not properly optimized; use automated evaluations to benchmark configurations and iterate until triggering is robust.