skills.rest
forgefabrik
forgefabrikCOMMUNITY
Explore all skills by this author →
GitHub

skill-creator

Drafts SKILL.md files, runs eval comparisons, and iteratively improves Claude skills.

Updated Jun 5, 2026
Search Tags:#prompt engineering#claude code#skill creation#benchmarking#evals#trigger optimization#package skill
One-click install
npx skills add https://github.com/forgefabrik/agent-bigbrother --skill skill-creator-forgefabrik
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/forgefabrik/agent-bigbrother/tree/main/.agents/skills/skill-creator
Command: npx skills add https://github.com/forgefabrik/agent-bigbrother --skill skill-creator-forgefabrik

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Skill Creator helps you turn an idea into a working Claude skill, then improve it with real examples instead of guesswork.

Core Features & Use Cases

  • Draft new SKILL.md files with clear triggering language and practical instructions.
  • Design realistic eval prompts, run comparisons, and inspect pass/fail patterns.
  • Improve existing skills by tightening the description, adding helpers, and packaging the final result.
  • Use it when you need to create a skill from scratch, tune trigger accuracy, benchmark changes, or prepare a distributable skill file.

Quick Start

Describe the skill you want to build or improve, and use Skill Creator to draft the skill, test it with representative prompts, review the results, and iterate until it works well.

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 test Claude skills with evals?▼

To create Claude skills, draft a SKILL.md file, design realistic eval prompts, run baseline comparisons, inspect pass/fail patterns, and iterate until trigger accuracy meets your requirements.

What is the best way to benchmark trigger accuracy for a Claude skill?▼

Benchmark trigger accuracy by running structured eval sets against your skill, comparing the results to a baseline, and inspecting pass/fail patterns to identify and refine ambiguous triggering language.

Can I package a finished Claude skill for distribution?▼

You can package a finished Claude skill by tightening the description, adding optional helper scripts and references for deterministic validation, and preparing the final distributable skill file.

How do I improve an existing Claude skill without guesswork?▼

Improve existing Claude skills by designing realistic eval prompts, running comparisons against a baseline, inspecting pass/fail patterns, and tightening the description or adding helpers based on real examples.

Do I need pyyaml to design and run Claude skill evals?▼

Yes, pyyaml is required as a dependency to support the scripts and references needed to run structured eval sets and perform deterministic validation workflows.

What should I include in a SKILL.md file for Claude?▼

A SKILL.md file should include clear triggering language, practical instructions for the task, and optional references to helper scripts that ensure deterministic validation and accurate execution.

Related Skills

Explore More Software Engineering Skills→
zhengfranzhengfran

skill-creator

Create and iteratively improve Claude Code skills with SKILL.md drafts and eval runs.

Community
Advanced
yonderlabyonderlab
1

skill-creator

Create standardized Skill Units packaged as distributable .skill files.

Official
Intermediate
GoldziherGoldziher
115

spikard-testing-strategy

Validate test fixtures for cross-language API handlers generated by Spikard.

Community
Intermediate
amhuppertamhuppert
2

skill-creator

Create and refine Claude Skill units with YAML frontmatter and Markdown instructions.

Community
Intermediate