skills.rest
archubbuck
archubbuckCOMMUNITY
Explore all skills by this author →
GitHub

skill-creator

Draft SKILL.md files, run evaluation loops, and benchmark trigger performance.

17|3|Updated Nov 21, 2025
Search Tags:#automation#workflow#evaluation#iteration#skill
One-click install
npx skills add https://github.com/archubbuck/workspace-architect --skill skill-creator-archubbuck
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/archubbuck/workspace-architect/tree/main/assets/skills/skill-creator
Command: npx skills add https://github.com/archubbuck/workspace-architect --skill skill-creator-archubbuck

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 creation and ongoing improvement is often a multi-step, error-prone process that requires coordinating drafts, tests, and refinements across teams. This Skill provides a structured workflow to create new skills, adjust existing ones, and measure performance with automated evaluations and benchmarks, enabling reliable, repeated improvements.

Core Features & Use Cases

  • Create new SKILL.md drafts and describe triggering rules, scope, and usage for a skill.
  • Run evaluation loops to test trigger accuracy, collect results, and identify edge cases.
  • Benchmark performance across configurations (with_skill vs without_skill) and compute delta metrics.
  • Iterate and optimize the skill description to improve triggering reliability and user outcomes.
  • Support for multi-domain workflows, reference materials, and helper scripts via bundled resources (scripts, references, assets).

Quick Start

Draft a new SKILL.md with a clear name and description, then run the evaluation loop to test and iterate.

Frequently Asked Questions about skill-creator

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

FAQPage Schema
How do I automate skill creation and evaluation workflows?▼

Automating skill creation involves drafting a SKILL.md file, running evaluation loops to test trigger accuracy, and iterating on the description to optimize performance. This structured workflow coordinates drafts, tests, and refinements to ensure reliable improvements.

What is the best way to benchmark skill performance across configurations?▼

Benchmarking skill performance is done by comparing with_skill and without_skill configurations to compute delta metrics. This automated evaluation collects results, identifies edge cases, and measures performance variations to guide optimization.

How do I test trigger accuracy for automation workflows?▼

Testing trigger accuracy requires running evaluation loops that validate triggering rules against test cases. This process collects results, identifies edge cases, and supports iterating improvements to optimize triggering reliability and user outcomes.

Do I need pyyaml and anthropic dependencies to iterate on skill descriptions?▼

Yes, pyyaml and anthropic dependencies are required to run the automated evaluation loops and benchmarking processes. These dependencies support validating triggers, computing delta metrics, and iterating improvements for skill optimization.

Can I use bundled scripts and references for multi-domain skill workflows?▼

Bundled scripts, references, and assets support multi-domain workflows by providing helper scripts and reference materials. These resources coordinate drafts, tests, and refinements across teams to enable reliable, repeated improvements for complex skill lifecycles.

Related Skills

Explore More Product & Management Skills→
coreyjacoreyja
12

skill-writer

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

Community
Basic
FxHollowFxHollow

canary

Monitor deployed URLs for console errors, performance regressions, and page failures.

Community
Advanced
ManuelKugelmannManuelKugelmann
3

claude-skill-creator

Create or update Claude skills with SKILL.md frontmatter and resource bundles.

Community
Intermediate
LuisSambranoLuisSambrano
7

llm-eval

Evaluate LLM applications using automated metrics and human feedback.

Community
Advanced