One-click install
npx skills add https://github.com/valorisa/Claude-Skills --skill skill-creator-valorisa
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/valorisa/Claude-Skills/tree/main/skills/skill-creator
Command: npx skills add https://github.com/valorisa/Claude-Skills --skill skill-creator-valorisa

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Writing custom Claude Code skills from scratch requires making dozens of small, easy-to-miss decisions about scope, triggers, and behavior, often leading to skills that don't match your original intent, require extensive rework, or become too complex to maintain.

Core Features & Use Cases

  • Guided 5-Question Interview: Extracts your skill's purpose, triggers, output expectations, boundaries, and tool needs one question at a time, proposing concrete interpretations for vague answers to avoid ambiguity.
  • Automated SKILL.md Generation: Builds a complete, structured skill file tailored to your answers, placed directly in your Claude Code skills directory with no manual formatting required.
  • Built-in Smoke Testing & Iteration: Immediately tests the generated skill with realistic prompts, shows you how it would respond, and refines it based on your feedback until it works exactly as you want.
  • Use Case Example: If you need a skill to convert technical API documentation into plain-language explanations for non-technical team members, the Skill Creator will help you define trigger phrases, output formatting rules, and edge cases to avoid, then validate the skill works correctly before you deploy it.

Quick Start

Use the skill-creator skill to build a custom Claude Code skill for summarizing GitHub pull request diffs by answering five short questions.

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 custom Claude Code skill without writing markdown from scratch?

To create a custom Claude Code skill without manual markdown formatting, use this guided interview process. It extracts your skill's purpose, triggers, and boundaries through five questions, then automatically generates a structured SKILL.md file tailored to your answers.

What is the best way to define scope and triggers for workflow automation commands?

The best way to define scope and triggers for workflow automation commands is via a guided five-question interview. This method proposes concrete interpretations for vague answers to eliminate scope ambiguity, ensuring your custom commands behave exactly as intended.

How does smoke testing work when generating custom commands for Claude Code?

Smoke testing for generated custom commands works by immediately running the new skill with realistic prompts within Claude Code. It shows you how the generated skill responds and allows you to iteratively refine the behavior based on your feedback until it works correctly.

Can I use guided skill creation to build code review and document processing tasks?

Yes, you can use guided skill creation to build code review and document processing tasks. The interview process supports building simple to moderately complex custom skills for these use cases by defining specific trigger phrases, output formatting rules, and edge cases to avoid.

Does generating a SKILL.md file require manual placement in the Claude Code skills directory?

Generating a SKILL.md file requires no manual placement in the Claude Code skills directory. The automated generation process builds the complete structured skill file and places it directly in the correct directory, ensuring immediate compatibility with Claude Code's skill loading system.

Why do my custom Claude Code skills fail to match my original intent?

Custom Claude Code skills fail to match your original intent due to dozens of small, easy-to-miss decisions about scope and behavior during manual creation. Using a guided interview process eliminates this manual effort and scope ambiguity, preventing extensive rework.