skills.rest
xtrm-dev
xtrm-devOFFICIAL
Explore all skills by this author →
GitHub

skill-creator

Draft, test, and refine AI skills with evals and benchmark comparisons.

4|1|Updated Nov 1, 2025
Search Tags:#claude code#benchmarking#skills#prompt design#evaluation#iteration#trigger optimization
One-click install
npx skills add https://github.com/xtrm-dev/specialists --skill skill-creator-xtrm-dev
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/xtrm-dev/specialists/tree/main/.xtrm/skills/default/skill-creator
Command: npx skills add https://github.com/xtrm-dev/specialists --skill skill-creator-xtrm-dev

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Skill Creator helps you design new skills, refine existing ones, and verify that they trigger and perform well instead of relying on guesswork.

Core Features & Use Cases

  • Skill Drafting: Turn a rough idea into a structured SKILL.md with clear triggering guidance and workflow instructions.
  • Evaluation and Iteration: Create realistic test prompts, run baseline comparisons, grade results, and improve the skill based on evidence.
  • Description Optimization: Tune the frontmatter description so the skill triggers more reliably for the right kinds of user requests.
  • Use Cases: Build a skill from scratch, repair a weak skill, benchmark trigger accuracy, or package a finished skill for reuse.

Quick Start

Tell the skill creator what you want the skill to do, and it will help you draft, test, evaluate, and iterate on the skill until it is ready.

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 test AI skills from scratch?▼

You can optimize skill triggering by tuning the frontmatter description, testing trigger accuracy against realistic prompts, and iterating until the skill reliably activates for the right user requests.

What's the best way to evaluate and iterate on prompt design?▼

Evaluate prompt design by generating realistic test prompts, running baseline comparisons, grading the outputs, and refining the skill description based on aggregated benchmark performance data.

How does benchmarking improve trigger accuracy for AI skills?▼

Benchmarking improves trigger accuracy by running evaluation orchestration and grading against test prompts, allowing you to measure performance and tune the description for reliable activation.

Do I need pyyaml and anthropic dependencies to create skills?▼

Yes, creating and testing skills requires the anthropic and pyyaml dependencies to support frontmatter parsing, eval orchestration, grading, and benchmark aggregation throughout the workflow.

Can I repair a weak skill that triggers incorrectly?▼

You can repair a weak skill by running baseline comparisons, grading its trigger accuracy, and optimizing the frontmatter description until it reliably triggers for the intended user requests.

Related Skills

Explore More Software Engineering Skills→
Project-N-E-K-OProject-N-E-K-O
2.4k

skill-creator

Create, structure, and package modular Claude skills with YAML frontmatter.

Official
Intermediate
GayanKavindaGayanKavinda

skill-creator

Create, revise, and benchmark Claude skills with SKILL.md files.

Community
Advanced
gesmith0606gesmith0606

skill-creator

Create, evaluate, and iteratively improve SKILL.md descriptions with Python scripts.

Community
Advanced
MCCTeamMCCTeam
2.1k

skill-creator

Create and iterate Claude skill drafts and evaluation workflows.

Official
Advanced