What is the Scoring Methodology skill?

It is a framework that allows Claude to apply consistent, weighted scoring to various technical audits like security, testing, architecture, and API reviews.

How are raw scores calculated in this framework?

Raw scores are assigned based on the worst severity found: Critical issues cap scores at 59, Major issues cap them at 79, and Minor issues reduce scores incrementally.

Can this skill be used for DORA metrics?

Yes, it includes a specific Git Health severity scale that is DORA-weighted to help identify blocks to elite delivery performance.

What triggers an automatic failure during an audit?

In security audits, any single Critical finding causes an automatic fail. For git health, the detection of secrets or credentials results in an automatic Critical flag.

What are the standard pass thresholds?

Most commands like code-review and testing require a score of ≥ 75, while security audits require a higher threshold of ≥ 80 due to the higher risk involved.

Scoring Methodology Framework

Name: Scoring Methodology Framework
Author: bailejl

bybailejl

0•

Security & Testing

Standardizes how Claude computes and interprets audit scores using weighted categories and severity-based thresholds.

The Scoring Methodology skill provides a unified framework for evaluating codebases and technical assets during AI-readiness audits. It establishes rigorous criteria for raw score assignment, weighted category calculations, and severity classification (Critical, Major, Minor), ensuring consistent results across security, architecture, and API reviews. By defining clear pass/fail thresholds and auto-fail conditions, this skill enables Claude to provide objective, data-driven assessments of a project's quality and risk profile, making it easier to prioritize remediation efforts.

Key Features

01Weighted category scoring for balanced technical assessments

02Standardized score computation procedures with DORA-weighted git health

03Severity-based raw score caps for Critical and Major violations

040 GitHub stars

05Predefined pass/fail thresholds for security, testing, and architecture

06Automatic fail conditions for critical security findings or exposed secrets

Use Cases

01Calculating technical debt and AI-readiness metrics for legacy systems

02Performing security audits with consistent risk-level classification

03Standardizing code review scores across diverse project repositories

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add bailejl/dev-plugins scoring-methodology

For use in Claude.ai and ChatGPT

Key Features

01Weighted category scoring for balanced technical assessments

02Standardized score computation procedures with DORA-weighted git health

03Severity-based raw score caps for Critical and Major violations

040 GitHub stars

05Predefined pass/fail thresholds for security, testing, and architecture

06Automatic fail conditions for critical security findings or exposed secrets

Use Cases

01Calculating technical debt and AI-readiness metrics for legacy systems

02Performing security audits with consistent risk-level classification

03Standardizing code review scores across diverse project repositories

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add bailejl/dev-plugins scoring-methodology

For use in Claude.ai and ChatGPT