skill-creator

Automate the full lifecycle of Claude Code skills from drafting to trigger optimization.

4|Updated Mar 19, 2026
One-click install
npx skills add https://github.com/ClubMediterranee/ai-core --skill skill-creator-clubmediterranee
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/ClubMediterranee/ai-core/tree/main/skills/skill-creator
Command: npx skills add https://github.com/ClubMediterranee/ai-core --skill skill-creator-clubmediterranee

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Automates the full lifecycle of Claude Code skills, guiding you from intent capture through drafting, evaluation, review, iteration, and trigger-description optimization so you can ship reliable skills faster.

Core Features & Use Cases

  • Guided Drafting: Elicit and document skill intent, scope, and edge cases to produce a solid SKILL.md.
  • End-to-End Eval Loop: Run evaluations, review outputs, and iterate based on feedback to improve activation accuracy.
  • Benchmark & Review: Integrated eval viewer and benchmarking to quantify performance and identify improvement opportunities.
  • Trigger Optimization: Automatically refine the skill's description to maximize correct activations.
  • Packaging & Reuse: Bundle skills into distributable artifacts for reuse across teams.

Quick Start

Draft a SKILL.md for your idea, run the eval-improve loop, and iterate until the trigger description is optimized.

Frequently Asked Questions about skill-creator

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

FAQPage Schema
How do I automate Claude Code skill development from drafting to optimization?

Automating Claude Code skill development involves capturing intent, drafting SKILL.md, running evaluations, and iterating on the trigger description to maximize activation accuracy and ship reliable skills faster.

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

Evaluating and benchmarking Claude Code skills requires running an end-to-end eval loop, reviewing outputs via an integrated viewer, and iterating based on feedback to quantify performance and identify improvement opportunities.

How does trigger description optimization work for Claude Code skills?

Trigger description optimization automatically refines the skill's description text across diverse use cases, ensuring the skill activates correctly when needed and minimizing false triggers during workflow automation.

Do I need PyYAML to create and test Claude Code skills?

Yes, PyYAML is a required dependency for the skill-creator workflow, supporting the parsing and generation of YAML-based configuration files needed for deterministic tasks, context loading, and output templating.

Can I bundle Claude Code skills with scripts and references for team reuse?

Yes, you can package skills into distributable artifacts by bundling optional scripts, references, and assets, enabling consistent reuse across teams and diverse automation workflows.