skills.rest
murphyjoseph
murphyjosephCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

skill-creator

Guide drafting, evaluating, and refining SKILL.md files for Claude skills.

Updated Mar 1, 2026
Search Tags:#automation#workflow#iteration#skill#claude-code#eval#skill-creator
One-click install
npx skills add https://github.com/murphyjoseph/dojo-kit --skill skill-creator-murphyjoseph
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/murphyjoseph/dojo-kit/tree/main/.agents/skills/skill-creator
Command: npx skills add https://github.com/murphyjoseph/dojo-kit --skill skill-creator-murphyjoseph

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Authors and teams frequently struggle to craft effective Claude skills that trigger reliably and scale with evaluation. This Skill Creator helps you design, test, and improve skills from scratch, ensuring a repeatable loop for iteration and feedback.

Core Features & Use Cases

  • Draft SKILL.md frontmatter and body to scaffold a new skill with clear entry points.
  • Run evaluation loops to test trigger accuracy, compare skill versions, and track progress over time.
  • Iterate with structured feedback history, enabling versioned improvements and reproducibility for teams.

Quick Start

Draft the initial SKILL.md, set up an eval plan, and start the first iteration of testing and refinement.

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 Claude skill with reliable triggering?▼

To create a Claude skill, you draft the SKILL.md frontmatter and body to scaffold clear entry points, then run evaluation loops to test and optimize trigger accuracy across various contexts.

What is the best way to iterate on Claude skills and track improvements?▼

The best way to iterate on Claude skills is using structured feedback history to run evaluation loops, enabling versioned improvements, performance benchmarking, and reproducibility for your team.

Do I need eval scaffolding to build and test Claude skills?▼

Yes, you need access to eval scaffolding and history tracking to benchmark performance, compare skill versions, and execute the iterative testing required for reliable skill development.

Can I use scripts and references to automate deterministic tasks in a Claude skill?▼

Yes, you can include optional scripts, references, and assets within your skill structure to execute deterministic tasks and store templates for automated workflow processing.

How does an evaluation loop improve Claude skill trigger accuracy?▼

An evaluation loop improves trigger accuracy by repeatedly testing the skill against varied inputs, measuring performance, comparing versions, and feeding structured feedback back into the refinement process.

Why does my Claude skill fail to trigger reliably across different contexts?▼

A skill fails to trigger reliably without iterative evaluation. Drafting clear SKILL.md entry points and running benchmark eval loops across contexts ensures consistent activation and performance.

Related Skills

Explore More Software Engineering Skills→
elijahmurrayelijahmurray
2

worktree-setup

Scan a project to generate a .worktree.json file for git worktree port configurations.

Community
Intermediate
quanganh208quanganh208

ck:skill-creator

Create or update Claude skills with eval-driven iteration and bundled resources.

Community
Advanced
linxuan-syslinxuan-sys
190

skill-creator

Create and iteratively optimize Claude Code skills with SKILL.md and eval workflows.

Community
Advanced
notwillknotwillk

evaluate-skill-quality

Evaluate skill outputs across prompts, baselines, and versions.

Community
Advanced