Can I call the judge skill directly?

No, the judge skill is designed to be invoked automatically by the cookoff or omakase-off processes during the evaluation phase.

What are the five scoring criteria?

The criteria are Fitness for Purpose, Justified Complexity, Readability, Robustness & Scale, and Maintainability.

How does the 'Fitness Gate' work?

If there is a score difference of 2 or more in the Fitness category between implementations, the one with the higher fitness wins immediately.

What is the difference between a Cookoff and an Omakase evaluation?

A Cookoff compares different implementations of the same design, while an Omakase evaluation compares entirely different approaches or designs.

What happens if an implementation gets a score of 1 in any category?

Any criterion scoring a 1 triggers a 'Critical Flaw' gate, which automatically eliminates that implementation from the selection process.

Code Implementation Judge

Name: Code Implementation Judge
Author: 2389-research

by2389-research

•

Security & Testing

Evaluates and scores multiple code implementations using a standardized five-criteria framework to identify optimal software solutions.

The Judge skill provides a rigorous, objective methodology for comparing different code implementations or architectural approaches within Claude Code. By systematically assessing fitness for purpose, complexity, readability, robustness, and maintainability, it eliminates subjective bias and ensures that chosen solutions meet high production standards. This skill is essential for 'cookoffs' and 'omakase' workflows where multiple variants of a solution must be weighed against each other using hard gates and feasibility checks.

Key Features

01Automated 'Hard Gate' checks to disqualify flawed implementations

02Structured winner selection with transparent rationale and trade-offs

03Standardized 5-criteria scoring framework for objective evaluation

04Deep feasibility analysis covering O(n) complexity and memory usage

058 GitHub stars

06Side-by-side scorecard generation for rapid implementation comparison

Use Cases

01Evaluating different architectural designs for scalability and robustness

02Standardizing code review criteria for AI-generated implementation variants

03Comparing multiple coding patterns to solve the same functional requirement

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add 2389-research/claude-plugins judge

For use in Claude.ai and ChatGPT

Download Skill