skills.rest
living-relation
living-relationCOMMUNITY
Explore all skills by this author →
GitHub

skill-creator

Create and iteratively improve Claude SKILL.md content with trigger and output evaluations.

Updated Aug 27, 2026
Search Tags:#prompt engineering#benchmarking#Claude Code#skill authoring#evals#trigger optimization
One-click install
npx skills add https://github.com/living-relation/tailormade --skill skill-creator-living-relation
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/living-relation/tailormade/tree/main/.agents/skills/skill-creator
Command: npx skills add https://github.com/living-relation/tailormade --skill skill-creator-living-relation

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and assets (resource) components.

What problem does it solve?

Skill-creator helps you design, refine, and validate Claude skills so they trigger at the right times and produce reliable, measurable results instead of guessing.

Core Features & Use Cases

  • Skill drafting and iteration: Turn your intended workflow into a usable SKILL.md and improve it based on what happens in practice.
  • Test-case creation and evaluation: Create realistic prompts, run them with and without the skill, and compare behavior.
  • Quantitative benchmarking: Produce pass-rate summaries, timing, and token usage to understand performance and variance.
  • Description optimization for triggering accuracy: Improve the skill description so Claude uses the skill more accurately (and avoids under-triggering).

Quick Start

Use the skill creator to draft a new skill for a task, then generate a small set of test prompts and run them to compare outputs with and without the skill.

Frequently Asked Questions about skill-creator

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

FAQPage Schema
How do I benchmark Claude Code skills to measure pass rates and token usage?▼

Skill-creator benchmarks Claude Code skills by orchestrating test sets and generating summaries with pass-rate, timing, and token usage statistics. It compares behavior by running test prompts with and without the skill.

Why does my Claude skill fail to trigger or activate when expected?▼

Under-triggering in Claude skills often stems from suboptimal description wording. Skill-creator runs trigger evaluations and guides description optimization to improve invocation accuracy and ensure the skill activates correctly.

How do I create and iterate on a SKILL.md file for prompt engineering workflows?▼

To create and iterate on a SKILL.md file, draft your intended workflow and refine it based on practical test outputs. Skill-creator enables iterative improvement of SKILL.md content using trigger and output evaluations.

Can I run qualitative and quantitative evaluations on Claude skills without external dependencies?▼

Yes, you can run qualitative and quantitative evaluations without external dependencies. Skill-creator satisfies deterministic evaluation loops by orchestrating test sets and generating benchmark summaries based on grader and analyzer outputs.

What is the best way to optimize skill descriptions for accurate Claude invocation?▼

The best way to optimize skill descriptions is through trigger evaluation. Skill-creator generates test prompts, analyzes under-triggering issues, and guides description wording changes to improve how accurately Claude invokes the skill.

What are the limitations of deterministic evaluation loops for skill benchmarking?▼

Deterministic evaluation loops are limited by grader and analyzer output quality. Skill-creator relies on these outputs to guide iteration, meaning inaccurate graders can produce misleading pass-rate summaries and poor optimization guidance.

Related Skills

Explore More Software Engineering Skills→
nuclianuclia
56

skill-creator

Create, refine, and evaluate AI skills with SKILL.md and test cases.

Official
Advanced
coreyjacoreyja
12

skill-writer

Draft SKILL.md frontmatter and structure guidance for Claude Code skills.

Community
Basic
Nuwanda04Nuwanda04

senior-prompt-engineer

Analyze prompts, evaluate RAG systems, and design agentic workflows.

Community
Advanced
CeeKay1901CeeKay1901

skill-creator

Draft, test, and refine reusable skills with evaluation and trigger optimization.

Community
Advanced