comprehensive-review-full-review

Comprehensive Review - Comprehensive Code Review Workflow

Skills Overview

Comprehensive Review is a multi-stage code review workflow that coordinates multiple specialized agents to perform comprehensive, multi-dimensional quality assessments of code, covering architecture, security, performance, testing, documentation, and more.

Applicable Scenarios

Comprehensive code check before deployment

Before releasing to production, perform a complete code quality, security, and performance check to ensure the code meets company standards and best practices.

Baseline assessment before code refactoring

Before undertaking large-scale refactoring or modernization, comprehensively assess the existing codebase for technical debt, architectural risks, and opportunities for improvement.

Enterprise-level project quality audit

Provide standardized code quality assessments for enterprise projects and generate a comprehensive review report with prioritized, actionable recommendations.

Core Features

Multi-stage review workflow

Execute reviews sequentially through four independent stages (code quality & architecture, security & performance, tests & documentation, best practices), with each stage’s results providing context for the next to ensure systematic and comprehensive coverage.

Configurable review modes

Support multiple review configuration options, including security-focus mode (--security-focus), performance-critical mode (--performance-critical), TDD review mode (--tdd-review), strict mode (--strict-mode), and metrics reporting (--metrics-report), among others.

Comprehensive report generation

Consolidate the results from all stages into a single executable report categorized by priority (P0–P3), including detailed analysis and remediation suggestions for security vulnerabilities, performance bottlenecks, test coverage, and architectural risks.

Frequently Asked Questions

What stages does the comprehensive code review workflow include?

The workflow is divided into four stages: the first stage reviews code quality and architecture; the second performs security and performance analysis; the third evaluates test coverage and documentation completeness; the fourth validates framework best practices and CI/CD processes. Each stage’s results are passed to subsequent stages to ensure continuity in the review.

How do I use different review configuration modes?

Specify the mode via command-line parameters, for example --security-focus prioritizes security vulnerabilities and OWASP compliance, --performance-critical focuses on performance bottlenecks and scalability issues, and --framework [name] applies best practices for specific frameworks (such as React, Spring, Django).

How are priorities classified in the review report?

The report categorizes issues into four priorities: P0 (must be fixed immediately) includes severe security vulnerabilities and data loss risks; P1 (fix before the next release) includes performance problems that affect user experience and missing critical tests; P2 (plan for the next iteration) includes non-critical optimizations and documentation improvements; P3 (backlog tracking) includes style issues and minor optimizations.