skill-creator

Create, modify, evaluate, and optimize AI skills with automated testing and benchmarking.

4.6k|301|Updated Jan 26, 2026
One-click install
npx skills add https://github.com/EvoScientist/EvoScientist --skill skill-creator-evoscientist
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/EvoScientist/EvoScientist/tree/main/EvoScientist/skills/skill-creator
Command: npx skills add https://github.com/EvoScientist/EvoScientist --skill skill-creator-evoscientist

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill streamlines the entire lifecycle of creating and improving AI skills, from initial concept to robust performance measurement, reducing the burden on developers and ensuring high-quality, discoverable skills.

Core Features & Use Cases

  • Skill Creation: Guides users through drafting new skills from scratch.
  • Skill Improvement: Iteratively refines existing skills based on evaluation feedback.
  • Performance Measurement: Runs automated evaluations and benchmarks skill performance.
  • Description Optimization: Fine-tunes skill descriptions for accurate triggering.
  • Use Case: A developer wants to create a new skill to summarize research papers. They use this Skill to draft the SKILL.md, write test cases, evaluate the results, and optimize the description until the AI reliably triggers the skill for relevant user queries.

Quick Start

Use the skill-creator to help me draft a new skill for summarizing legal documents.

Frequently Asked Questions about skill-creator

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

FAQPage Schema
How do I build an AI skill from scratch?

To build an AI skill from scratch, you draft the initial SKILL.md structure, define core processing logic, and iteratively refine descriptions until the agent reliably triggers desired actions. This provides a structured skill creation framework.

How does automated benchmarking evaluate AI agent performance?

Automated benchmarking evaluates AI agent performance by running structured test cases against the skill, measuring output accuracy, and validating trigger reliability. This provides quantitative performance metrics for skill optimization.

What is the best way to optimize AI skill descriptions for accurate triggering?

The best way to optimize AI skill descriptions is through iterative description tuning based on evaluation feedback. This aligns skill metadata with user queries, ensuring accurate and reliable triggering by the AI agent.

Can I use this to iteratively refine an existing prompt engineering design?

Yes, you can iteratively refine existing prompt engineering designs by applying automated evaluation feedback. This identifies performance gaps and modifies skill logic to improve overall agent reliability and task execution.

Do I need to write test cases manually for AI skill evaluation?

No, you do not need to write test cases manually for AI skill evaluation. The skill provides tools to generate and run automated evaluations, benchmarking performance outputs directly without manual test setup.