skills.rest
cptunderpantsmoons
cptunderpantsmoonsCOMMUNITY
Explore all skills by this author →
GitHub

skill-creator

Automate Claude skill creation from draft to deployment with evaluation and iteration.

Updated Aug 23, 2026
Search Tags:#workflow#cli#evaluation#iteration#skill#project-management#creator
One-click install
npx skills add https://github.com/cptunderpantsmoons/contract-hubv1 --skill skill-creator-cptunderpantsmoons
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/cptunderpantsmoons/contract-hubv1/tree/main/.claude/skills/skill-creator
Command: npx skills add https://github.com/cptunderpantsmoons/contract-hubv1 --skill skill-creator-cptunderpantsmoons

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

The Skill Creator helps you compose, evaluate, and continuously improve Claude skills—from initial drafting to deployment—so you can build reliable, reusable capabilities quickly.

Core Features & Use Cases

  • Initialize new skills and register them into a global registry using the two-step workflow.
  • Iteratively evaluate prompts, run benchmarks, and refine the triggering description to maximize accuracy and coverage.
  • Manage evaluation loops, generate reports, and synthesize improvements across many prompts and scenarios.

Quick Start

Draft your skill, then initialize it with the skills tool, register it, and begin iterative evaluation to refine prompts, tests, and the triggering 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 iterate on Claude skills from a draft to deployment?▼

To create Claude skills, you draft the prompt, initialize it with the skills tool, register it globally, and run iterative evaluations to refine prompts and triggering descriptions for deployment.

What is the best way to test and evaluate prompts for a new skill?▼

The best way to evaluate prompts for a new skill is to run benchmark evaluations and iterate on the triggering description to maximize accuracy and coverage across various scenarios.

How does the skill initialization and registration workflow operate?▼

The skill initialization workflow operates through a two-step process using the skills tool to set up the skill structure and then register it into a global registry for access.

Do I need PyYAML to build and manage skills with this workflow?▼

Yes, you need PyYAML installed because the skill workflow relies on parsing a SKILL.md frontmatter file containing the name and description for proper initialization and registration.

Can I use scripts and references directories when structuring a project skill?▼

Yes, you can use optional scripts and references directories to organize your project skill, allowing you to manage evaluation loops and generate reports for continuous improvement.

Why does my Claude skill fail to activate even after initialization?▼

Your Claude skill may fail to activate if the triggering description in the SKILL.md frontmatter is not refined through iterative evaluation to maximize accuracy and coverage.

Related Skills

Explore More Product & Management Skills→
tarrragontarrragon

project-init

Validate OS, Python, UV, ripgrep, and project dependencies for environment setup.

Community
Advanced
TTAWDTTTTAWDTT
4

anthropic-skill-creator

Run structured evaluations and automated improvements on Claude-based skill descriptions.

Community
Advanced
FredericoMatsuokaFredericoMatsuoka

opensquad-skill-creator

Create, modify, and benchmark Opensquad skills with SKILL.md generation.

Community
Advanced
oabdelmaksoudoabdelmaksoud
4

skill-creator

Draft, test, evaluate, and iterate on OpenClaw SKILL.md-based skills.

Community
Advanced