skills.rest
openqa-labs
openqa-labsOFFICIAL
Explore all skills by this author →
WebsiteGitHub

skill-creator

Build, validate, and package Claude Code skills into .skill files.

Updated Apr 7, 2026
Search Tags:#automation#workflow#evaluation#iteration#skill#claude-code
One-click install
npx skills add https://github.com/openqa-labs/zenith --skill skill-creator-openqa-labs
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/openqa-labs/zenith/tree/main/.agents/skills/skill-creator
Command: npx skills add https://github.com/openqa-labs/zenith --skill skill-creator-openqa-labs

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill enables teams to design, test, and iteratively improve Claude Code skills end-to-end, reducing time-to-value and increasing predictability of skill triggers.

Core Features & Use Cases

  • End-to-end skill lifecycle: draft a skill, run evals, analyze results, rewrite the SKILL.md, and package into a .skill file.
  • CI-like validation: every commit can trigger validation and sanity checks before merging or publishing.
  • Opinionated defaults and tooling: guidance on structure, references, assets, and scripts for deterministic tasks.
  • Use Case: a team wants to build a new skill from scratch, iterate on its frontmatter description, and compare against baselines.

Quick Start

Use the Skill Creator to start a new skill from scratch, run the evaluation loop, and iterate until the results stabilize.

Frequently Asked Questions about skill-creator

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

FAQPage Schema
How do I build and validate Claude Code skills end-to-end?▼

To build and validate Claude Code skills, use an opinionated workflow that drafts the skill, runs YAML frontmatter validation, organizes scripts, references, and assets, and packages the output into a .skill file for distribution.

What is the best way to iterate on a Claude Code skill workflow?▼

The best way to iterate on a Claude Code skill workflow is to run an evaluation loop that tests the skill, analyzes results against baselines, and rewrites the SKILL.md until trigger predictability and performance stabilize.

Does Claude Code skill development support CI-like validation checks?▼

Yes, Claude Code skill development supports CI-like validation checks by triggering YAML frontmatter validation and structural sanity checks on every commit before merging or publishing the packaged .skill file.

How do I package a Claude Code skill for distribution?▼

To package a Claude Code skill for distribution, compile your validated SKILL.md along with its multi-resource directories (scripts, references, assets) into a single .skill file.

Can I organize scripts and references within a single Claude Code skill?▼

Yes, you can organize scripts, references, and assets within a single Claude Code skill by using an opinionated multi-resource folder structure to handle deterministic tasks safely.

Why does my Claude Code skill trigger unpredictably during evaluation?▼

Claude Code skills trigger unpredictably when the frontmatter description lacks precision, requiring iterative rewriting and evaluation loop analysis to stabilize trigger predictability against baseline metrics.

Related Skills

Explore More Product & Management Skills→
ResearAIResearAI
3.3k

baseline

Standardize a reference object for quest evaluation with plan and checklist requirements.

Community
Advanced
danielscholldanielscholl
23

test-framework

Validate plugin skills across four layers with dry-run and JSON CI outputs.

Community
Advanced
stravineststravinest

skill-creator

Create and maintain Claude skills with standardized SKILL.md templates and packaging.

Community
Intermediate
camoneartcamoneart
5

Creating Agent Skills

Guide Claude through creating structured Agent Skills with SKILL.md files.

Community
Basic