skill-creator

Define and bootstrap new Claude skills with SKILL.md and eval prompts.

1|Updated Jun 3, 2020
One-click install
npx skills add https://github.com/LanceRabbit/dotfiles --skill skill-creator-lancerabbit
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/LanceRabbit/dotfiles/tree/main/.claude/skills/skill-creator
Command: npx skills add https://github.com/LanceRabbit/dotfiles --skill skill-creator-lancerabbit

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill enables users to systematically create and iteratively improve Claude skills, including drafting SKILL.md, designing test prompts, running evals, and applying feedback to enhance triggering accuracy and performance.

Core Features & Use Cases

  • Draft and structure new skills with a clear frontmatter and instructions that load into context.
  • Design, run, and analyze eval prompts to iterate on triggering accuracy and task performance.
  • Scaffold bundled resources (scripts, references, assets) and manage versioned iterations for skill improvements.

Quick Start

Draft a new skill by outlining its intent, write the SKILL.md frontmatter, create a few test prompts, and iterate based on feedback.

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 bootstrap a new Claude skill from scratch?

To create a Claude skill, draft its intent and scope, write the SKILL.md with proper frontmatter, and structure the body instructions to load into context for production readiness.

How can I evaluate and iterate on prompt engineering for skill triggering accuracy?

Evaluate prompt engineering by designing test prompts, running evals, and applying quantitative and qualitative feedback to iterate on the skill's triggering accuracy and task performance.

What is the best way to structure a SKILL.md file for production readiness?

The best way to structure a SKILL.md is to enforce frontmatter, define the body instructions, manage dependencies, track components, and document safety considerations for a production-ready unit.

Can I bundle scripts and references when I create a new skill?

Yes, you can scaffold bundled resources like scripts, references, and assets, managing versioned iterations to systematically improve the skill across multiple domains.

What safety considerations do I need to document when creating a skill?

Creating a skill requires tracking dependencies, managing component interactions, and documenting safety considerations to ensure the final production-ready unit operates within defined boundaries.

Why should I use evals for skill creation instead of just writing documentation?

Using evals during skill creation captures evaluation criteria and applies feedback to iteratively enhance performance, ensuring the documentation translates to accurate triggering and execution.