How is the final grade calculated?

The grade is a weighted average: Security (25%), Code Quality (20%), Test Coverage (20%), API Compliance (20%), and UI Compliance (15%).

What is Agent Teams mode in this skill?

Agent Teams mode allows specialized sub-agents to share findings in real-time, such as a security auditor alerting the test generator about untested vulnerable paths.

Can I customize the verification scope?

Yes, the skill uses an interactive prompt to let you choose between full verification, security audits, test-only checks, or quick quality lints.

Does it track historical data?

Yes, it utilizes Memory MCP to store and query verification metrics, allowing you to track project quality trends over time.

What does the 'verify' skill actually do?

It runs a multi-phase verification process using specialized agents to evaluate security, quality, and tests, providing a detailed grade and actionable report.

Comprehensive Code Verification

Name: Comprehensive Code Verification
Author: yonatangross

byyonatangross

•

Seguridad y Pruebas

Verifies code implementations using parallel specialized agents and nuanced grading to ensure production readiness and security.

The Verify skill is a sophisticated orchestration layer for Claude Code that automates the validation of software changes. By deploying a team of five specialized AI agents—focusing on code quality, security, test coverage, backend architecture, and frontend standards—it provides a multi-dimensional analysis of any feature or implementation. It calculates a nuanced 0-10 composite score, generates prioritized improvement suggestions based on effort/impact ratios, and tracks historical metrics via memory to ensure consistent quality standards across the development lifecycle.

Características Principales

01Interactive scope selection for targeted audits or full-stack verification

02Historical metrics tracking and trend analysis via Memory MCP

03Nuanced 0-10 grading system with composite scoring and letter grades

04Parallel multi-agent analysis (Security, Quality, Backend, Frontend, Tests)

0569 GitHub stars

06Effort vs. Impact prioritized improvement suggestions for quick wins

Casos de Uso

01Validating a new feature implementation before merging a pull request

02Benchmarking code quality and test coverage against project-specific policies

03Performing automated security and architectural audits on legacy codebases

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add yonatangross/orchestkit verify

For use in Claude.ai and ChatGPT

Download Skill