skill-creator

Create, modify, and optimize AI skills with performance measurement.

4|Updated Jun 16, 2026
One-click install
npx skills add https://github.com/zhu327/pi --skill skill-creator-zhu327
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/zhu327/pi/tree/main/.pi/agent/skills/skill-creator
Command: npx skills add https://github.com/zhu327/pi --skill skill-creator-zhu327

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires eval-viewer/generate_review.py, scripts/aggregate_benchmark.py, scripts/run_loop.py, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill provides a comprehensive toolset for creating, modifying, and optimizing AI skills, along with measuring their performance.

Core Features & Use Cases

  • Skill Creation: Develop new skills from scratch, define their behavior, and integrate them into the system.
  • Skill Modification: Refine existing skills based on feedback and evaluation results.
  • Performance Measurement: Benchmark skill performance with variance analysis and iterate to improve accuracy and efficiency.
  • Use Case: Imagine you want to create a skill to analyze financial data. Use this Skill to design the skill, write test cases, evaluate its performance, and optimize its description for better triggering accuracy.

Quick Start

Use the skill-creator skill to create a new skill for financial data analysis.

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 optimize AI skills with performance measurement?

To create and optimize AI skills, you define the skill behavior, integrate it into the system, and use variance analysis to benchmark performance for better triggering accuracy.

What is the best way to measure AI skill performance and improve accuracy?

Measuring AI skill performance involves running benchmarks with variance analysis and iterating on the feedback to optimize skill descriptions and improve overall triggering accuracy.

Can I modify existing AI skills based on evaluation results?

Yes, you can modify existing AI skills by refining their behavior based on direct feedback and evaluation results to enhance their efficiency and accuracy within the system.

Do I need a well-defined skill structure to benchmark performance with variance analysis?

Yes, a well-defined skill structure is required to benchmark performance with variance analysis, ensuring the evaluation tools can accurately measure and optimize the AI skill behavior.

What tools are needed to design and evaluate new AI skills?

Designing and evaluating new AI skills requires access to performance measurement tools, including scripts for running evaluation loops, aggregating benchmarks, and generating reviews.