skill-create

Create SKILL.md frontmatter, Markdown instructions, and optional resources for Claude Skills.

Updated Mar 23, 2026
One-click install
npx skills add https://github.com/mia-cx/.agents --skill skill-create-mia-cx
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-create
Source: https://github.com/mia-cx/.agents/tree/main/skills/skill-create
Command: npx skills add https://github.com/mia-cx/.agents --skill skill-create-mia-cx

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

The Skill Creator streamlines the end-to-end process of designing, validating, and iterating Claude skills from initial concept to deployment, providing a repeatable workflow, templates, and tooling to accelerate skill maturity.

Core Features & Use Cases

  • Capture intents and draft SKILL.md with a clear triggering description.
  • Build and run a structured eval loop (prompts, baseline comparisons, and iterative improvements).
  • Manage bundled resources (scripts, references, assets) to support reliable and measurable skills.
  • Generate, store, and review test results, metrics, and feedback to drive data-informed iteration.

Quick Start

Draft an initial SKILL.md and run the evaluation loop to begin iterating on triggers.

Frequently Asked Questions about skill-create

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

FAQPage Schema
How do I create and structure a SKILL.md file for Claude prompting?

To create a SKILL.md file, you draft frontmatter for metadata and a Markdown body for operational instructions. This structure ensures triggering conditions, compatibility, and usage patterns are documented to facilitate reliable skill activation.

What is the best way to evaluate and benchmark Claude skills?

Evaluating Claude skills requires a structured eval loop with prompts, baseline comparisons, and improvement iterations. This process generates test results and metrics to drive data-informed iterations and benchmark skill reliability.

How do I bundle scripts and references into a custom skill?

You bundle resources by loading optional scripts, references, and assets into the skill directory. Managing these bundled components ensures the skill has the necessary external tooling and context to support reliable execution.

Do I need YAML to manage Claude skill creation?

YAML is required to structure the SKILL.md frontmatter. This dependency allows you to define the skill's triggering description and metadata needed for successful discovery and activation.

How do I iterate on a skill prompt that is not triggering correctly?

When a skill prompt is not triggering correctly, you run the structured evaluation loop to review test results and feedback. This data-driven iteration helps refine the triggering description and improve activation accuracy.

Can I use this workflow to test skill compatibility and usage patterns?

Yes, the skill creation workflow explicitly validates compatibility and usage patterns. Documenting these elements during the design phase ensures the skill operates reliably within the intended environment.