skills.rest
jie-meng
jie-mengCOMMUNITY
Explore all skills by this author →
GitHub

skill-creator

Create and iteratively improve AI skills and prompts across multiple LLM platforms.

1|Updated Mar 12, 2019
Search Tags:#automation#benchmarking#packaging#claude-code#prompt-engineering#skill-creator#evals
One-click install
npx skills add https://github.com/jie-meng/toolscripts --skill skill-creator-jie-meng
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/jie-meng/toolscripts/tree/main/ai/skills/skill-creator
Command: npx skills add https://github.com/jie-meng/toolscripts --skill skill-creator-jie-meng

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Streamlines the creation, evaluation, and iterative improvement of skills and system prompts for multiple AI platforms so teams can produce reliable, testable skills without reinventing infrastructure for every iteration.

Core Features & Use Cases

  • Multi-platform authoring: Draft SKILL.md frontmatter and instruction bodies for Claude Code, Claude.ai, OpenAI/ChatGPT, Cursor, Windsurf, and generic .md skill formats.
  • Evaluation & benchmarking: Generate eval sets, run parallel with-skill and baseline runs, grade outputs, aggregate statistics, and produce human-facing benchmark reports.
  • Improve & package: Automatically propose description improvements, run optimization loops, and package validated skills into distributable .skill archives using included scripts.
  • Use Case: Rapidly create a new Claude Code skill, run trigger evaluations versus baselines, iterate on descriptions and assertions, and produce a benchmarked, packaged skill ready for deployment.

Quick Start

Ask the assistant to "Help me create a Claude Code SKILL.md for [task], draft 3 test prompts, and run a baseline vs with-skill evaluation."

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 evaluate SKILL.md prompts across multiple AI platforms?▼

You can create and evaluate AI skills across platforms like Claude Code, OpenAI, and Cursor by drafting SKILL.md frontmatter, generating evaluation sets, and running benchmark comparisons to iteratively improve system prompts.

What is the best way to benchmark system prompts against baselines?▼

Benchmarking system prompts involves generating eval sets, running parallel baseline and with-skill evaluations, grading outputs, and aggregating statistics into human-facing benchmark reports to measure skill performance.

Does this skill support packaging AI prompts for deployment?▼

Yes, skill packaging is supported through included scripts that validate skills and bundle them into distributable .skill archives ready for deployment across different LLM platforms.

Can I use prompt engineering to optimize trigger evaluations for Claude Code?▼

Yes, you can optimize Claude Code skills by drafting trigger evaluations, running optimization loops on descriptions and assertions, and producing benchmarked skills ready for deployment.

Do I need anthropic and pyyaml dependencies to run skill evaluations?▼

Yes, the skill requires the anthropic and pyyaml dependencies to draft SKILL.md metadata, run trigger and baseline evals, and aggregate grader outputs into benchmark artifacts.

Why do my AI skill evaluations fail to aggregate grader outputs?▼

Aggregating grader outputs fails when evaluation sets are misaligned or dependencies are missing, preventing the generation of accurate benchmark artifacts and human-facing reports.

Related Skills

Explore More Software Engineering Skills→
modu-aimodu-ai
1.2k

moai-ai-nano-banana

Automate content and image generation via Nano Banana Pro AI.

Official
Advanced
I194I194
9

canary

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

Community
Advanced
dakesandakesan
2

skill-creator

Create or update Claude Skills with YAML frontmatter and SKILL.md content.

Community
Basic
mcauduro0mcauduro0
1

senior-prompt-engineer

Analyze and optimize LLM prompts with Python scripts.

Community
Advanced