skill-creator

Create, modify, and optimize AI skills with test cases and performance evaluation.

1.1k|142|Updated Sep 4, 2024
One-click install
npx skills add https://github.com/Onelevenvy/flock --skill skill-creator-onelevenvy
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/Onelevenvy/flock/tree/main/flock-data/skills/skill-creator
Command: npx skills add https://github.com/Onelevenvy/flock --skill skill-creator-onelevenvy

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires subagents, evaluation framework, performance analysis tools, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill addresses the complexities of skill development, from creating new skills to modifying existing ones and optimizing their performance.

Core Features & Use Cases

  • Skill Creation: Facilitates the creation of new skills from scratch.
  • Skill Modification: Allows for editing and improving existing skills.
  • Skill Optimization: Provides tools for benchmarking skill performance and optimizing skill descriptions for better triggering accuracy.
  • Use Case: If you need to create a skill that can automatically generate data visualizations from CSV files, this Skill can guide you through the process, from defining the skill's intent to optimizing its description for better triggering accuracy.

Quick Start

Use the skill-creator skill to define the purpose of your new skill, and I'll help you create it step by step.

Frequently Asked Questions about skill-creator

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

FAQPage Schema
How do I create a new AI skill from scratch?

To create a new AI skill from scratch, define the skill's purpose and intent, then use the framework to generate the definition, test cases, and optimized descriptions step by step.

How do I optimize AI skill descriptions for better triggering accuracy?

Optimize AI skill descriptions by using built-in performance evaluation and benchmarking tools to analyze triggering behavior, then refining the description text to improve accuracy.

Do I need subagents to run performance benchmarking on AI skills?

Yes, subagents are required for performance benchmarking and analysis, as they enable the parallel execution needed to evaluate test cases and measure skill behavior effectively.

Can I modify existing AI skills to support web browsing and command execution?

Yes, you can modify existing AI skills to support various types including web browsing, command execution, and file processing by editing the skill definition within the framework.

What is the best way to benchmark AI skill performance?

The best way to benchmark AI skill performance is through the integrated evaluation framework, which handles test case creation and performance analysis to measure execution results.