code-review

Perform multi-model codebase reviews with severity findings and scorecards.

Updated Jan 15, 2026
One-click install
npx skills add https://github.com/venturecrane/crane-console --skill code-review-venturecrane
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: code-review
Source: https://github.com/venturecrane/crane-console/tree/main/.agents/skills/code-review
Command: npx skills add https://github.com/venturecrane/crane-console --skill code-review-venturecrane

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Codebase Review enables teams to perform thorough, multi-model analyses of a codebase, producing a graded scorecard stored in VCMS and a full report committed to the repository.

Core Features & Use Cases

  • Multi-model perspectives: evaluates architecture, security, code quality, testing, dependencies, documentation, and Golden Path compliance.
  • Graded outputs: generates a formal scorecard in VCMS and a comprehensive repository report.
  • Actionable recommendations: surfaces concrete improvements with prioritized remediation guidance.

Quick Start

Run a full codebase review on the target repository and review results in VCMS.

Frequently Asked Questions about code-review

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I perform a multi-model codebase review to identify architectural risks?

You can run a full codebase review on the entire repository or target a focused path within the codebase, enabling automated context extraction to surface concrete improvement opportunities with prioritized remediation guidance and a graded scorecard.

Can I run a quick codebase analysis on just a focused path within my repository?

Quick codebase analysis on a focused path supports Claude-only reviews for faster iteration, while full multi-model analyses evaluate Golden Path compliance, security, and testing across the broader codebase.

What does a codebase review scorecard include for tracking code quality?

The scorecard grades your codebase across multiple dimensions and stores results in VCMS, while the full report committed to your repository details structured findings with severity and prioritized remediation guidance.

Does automated codebase analysis work without external dependencies?

The codebase analysis operates with zero external dependencies, relying on automated context extraction and multi-model insights to generate structured findings and a formal scorecard.

When should I use a full multi-model codebase analysis instead of a quick review?

A full multi-model analysis provides comprehensive coverage of dependencies, documentation, and code quality, while quick reviews offer faster iteration for immediate codebase feedback.