How is the Code Quality Score determined?

The score starts at 100 points. Penalties are subtracted based on metric violations (e.g., -10 for high complexity) and issue severity (High: -20, Medium: -10, Low: -3).

Does this skill automatically fix my code?

No, it serves as an automated reviewer. it provides a detailed verdict and list of issues but never modifies your source files or project management status directly.

What specific code metrics does this skill track?

The skill monitors cyclomatic complexity (target ≤10), function size (≤50 lines), file size (≤500 lines), nesting depth (≤3), and parameter counts (≤4).

Can I run the checker without using external research tokens?

Yes, you can use the --skip-mcp-ref flag to enable fast-track mode, which focuses on local static analysis and metrics to reduce token usage.

What is the benefit of MCP Ref validation?

It allows the skill to perform external research to ensure your implementation uses the most optimal approach and follows the latest best practices for your specific tech stack.

Code Quality & Architecture Checker

Name: Code Quality & Architecture Checker
Author: levnikolaevich

bylevnikolaevich

•

147

•

Seguridad y Pruebas

Performs automated code quality audits using quantitative metrics, architectural validation, and MCP-driven best practice research.

The ln-511 Code Quality Checker provides a rigorous, objective quality gate for your development workflow. It calculates a comprehensive Code Quality Score by evaluating cyclomatic complexity, nesting depth, and file sizes while penalizing architectural anti-patterns. Unlike standard linters, this skill leverages MCP (Model Context Protocol) to perform deep research into the optimality of your chosen implementation patterns, checking for performance bottlenecks, security vulnerabilities, and adherence to DRY/KISS/YAGNI principles. It is the essential tool for maintaining high standards in production-ready codebases.

Características Principales

01Automated detection of layer boundary violations, God classes, and side-effect breadth.

02Quantitative Code Quality Scoring (0-100) based on metric penalties and issue severity.

03Structured issue reporting with standardized prefixes like SEC-, PERF-, ARCH-, and MNT-.

04147 GitHub stars

05MCP-driven validation of architectural optimality, performance, and best practices.

06Cross-story analysis to identify duplicate logic (DRY) and unused exports across the codebase.

Casos de Uso

01Validating new library implementations against current industry best practices.

02Conducting architectural audits to identify technical debt and structural drift.

03Establishing automated quality gates for feature branches before final review.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add levnikolaevich/claude-code-skills ln-511-code-quality-checker

For use in Claude.ai and ChatGPT

Download Skill