How does the scoring system work?

Each dimension is rated from 0 to 10 based on a standardized rubric. A weighted average is calculated to provide an overall grade ranging from A+ (Excellent) to F (Critical).

Can I use this for things other than source code?

Yes, the skill is designed to assess code, design documents, database schemas, caching strategies, or general technical approaches.

Does it provide actionable feedback?

Yes, it generates a prioritized list of improvement suggestions categorized by impact and effort, specifically highlighting 'Quick Wins' that can be implemented in under an hour.

What is the benefit of 'Agent Teams' mode during assessment?

In Agent Teams mode, specialized AI agents (e.g., a security auditor and a performance engineer) cross-validate each other's findings and discuss disagreements to ensure a more accurate and balanced score.

Assessment & Quality Auditor

Name: Assessment & Quality Auditor
Author: yonatangross

byyonatangross

•

Security & Testing

Evaluates code, designs, and technical approaches using a structured 0-10 scoring system with detailed pros, cons, and improvement plans.

The Assess skill provides a comprehensive framework for determining the quality of any development artifact, from source code to architectural patterns. By analyzing six core dimensions—correctness, maintainability, performance, security, scalability, and testability—it generates a weighted composite score and a letter grade. It features multi-agent orchestration to cross-validate findings, delivering a deep-dive report that includes prioritized 'Quick Wins', alternative comparisons, and effort estimates for suggested improvements. This tool is essential for teams needing objective, high-fidelity technical reviews during the development lifecycle.

Key Features

01Prioritized improvement roadmap identifying high-impact 'Quick Wins'

02Six-dimension quality scoring (0-10) with weighted composite results

03Detailed pros/cons analysis with impact and severity evidence

0469 GitHub stars

05Multi-agent team orchestration for cross-validated security and performance audits

06Comparative matrix for evaluating the current approach against alternatives

Use Cases

01Comparing multiple technical implementation strategies to determine the best trade-offs

02Performing deep-dive code reviews for critical backend services or security modules

03Evaluating architectural design documents and database schemas before implementation

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add yonatangross/orchestkit assess

For use in Claude.ai and ChatGPT

Download Skill