skills.rest
dduquenne
dduquenneCOMMUNITY
Explore all skills by this author →
GitHub

skill-creator

Create and iterate Claude skills with SKILL.md files and evaluation loops.

1|Updated Mar 10, 2026
Search Tags:#automation#qa#evaluation#iteration#skill#prompt-engineering
One-click install
npx skills add https://github.com/dduquenne/unanima-platform --skill skill-creator-dduquenne
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/dduquenne/unanima-platform/tree/main/.claude/skills/skill-creator
Command: npx skills add https://github.com/dduquenne/unanima-platform --skill skill-creator-dduquenne

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

The Skill Creator provides a repeatable workflow for designing, evaluating, and iterating Claude skills—from drafting the SKILL.md entry point to running quantitative evaluations, analyzing results, and rewriting the skill based on feedback to improve triggering accuracy and performance.

Core Features & Use Cases

  • End-to-end workflow: draft a skill, run evals, analyze results, and iterate the skill content and tests.
  • Built-in resources: access scripts, references, and assets to support deterministic tasks, deep dives, and output templates.
  • Benchmarking and packaging: generate benchmarks to compare iterations and package skills as distributable artifacts for distribution.

Quick Start

Create an initial SKILL.md for your idea, then run the built-in evaluation loop to iterate until triggering and performance are stable.

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 iterate Claude skills with automated evaluation?▼

Create and iterate Claude skills by drafting a frontmatter-driven SKILL.md entry point, then running a built-in evaluation loop to analyze triggering accuracy and rewrite skill content based on feedback. This workflow consolidates end-to-end skill creation and optimization.

What is the best way to benchmark and optimize prompt engineering for skill triggering?▼

Benchmark and optimize prompt engineering by running quantitative evaluations on your skill descriptions, analyzing the results, and generating benchmarks to compare iterations until triggering accuracy and performance are stable.

Can I use this skill creation workflow to package and distribute my automation assets?▼

Yes, you can package skills as distributable artifacts. The workflow provides scripts, references, and output templates to support deterministic tasks and allows you to generate distributable artifacts after benchmarking.

Do I need pyyaml to run the skill evaluation loop?▼

Yes, pyyaml is a required dependency. The skill relies on it to parse the frontmatter-driven entry point and manage the configuration needed for the evaluation and iteration loop.

How does the evaluation loop work when refining a SKILL.md file?▼

The evaluation loop works by running quantitative tests against your skill, analyzing the results to identify triggering or performance issues, and rewriting the skill content and tests based on that feedback to improve accuracy.

Related Skills

Explore More Software Engineering Skills→
eduardoabreu81eduardoabreu81
6

template-skill

Create a standardized SKILL.md structure with name and description frontmatter fields.

Community
Basic
lDuanglDuang

skill-creator

Create and refine Claude Skills with SKILL.md and trigger-eval loops.

Community
Advanced
SJTU-IPADSSJTU-IPADS
7

senior-prompt-engineer

Design, analyze, and optimize prompts with evaluation frameworks and structured outputs.

Official
Advanced
easyspace-aieasyspace-ai

skill-creator

Create, evaluate, and optimize Claude Skills with templates and benchmark loops.

Community
Advanced