Code Quality Review

Code Quality Review

Overview

Conducts systematic code quality analysis across multiple dimensions: maintainability, readability, complexity, design patterns, naming conventions, code duplication, and adherence to best practices. Produces actionable feedback with severity ratings and specific improvement recommendations.

Core Capabilities

1. Code Smells Detection - Identifies bloaters, object-orientation abusers, change preventers, dispensables, and couplers
2. Complexity Analysis - Measures cyclomatic and cognitive complexity with risk assessment
3. Maintainability Assessment - Evaluates code maintainability index and technical debt
4. Design Pattern Evaluation - Reviews architectural patterns and SOLID principles
5. Best Practices Validation - Checks adherence to language-specific standards and conventions

Review Workflow

Step 1: Scope Assessment

Determine review scope based on change size:

• Small (500 lines): Architectural review, break into smaller reviews if possible, 2-4 hours

For scope-specific guidance, see [review-scope-guidelines.md](references/review-scope-guidelines.md)

Step 2: Initial Assessment

Gather Context:

• Identify programming language and framework
• Understand project type (web app, API, library, CLI, etc.)
• Note existing coding standards or style guides
• Check for linter configuration files (.eslintrc, .pylintrc, checkstyle.xml, etc.)

Read the Code:

• Start with entry points (main files, index files)
• Review module/package organization
• Check dependency management
• Examine test files if available

Step 3: Quality Analysis

Analyze code across key dimensions:

• Code Smells: Long methods, large classes, duplicate code, dead code, etc.
• Complexity: Cyclomatic complexity (target <15), cognitive complexity, nesting depth
• Maintainability: Clear naming, proper abstraction, separation of concerns
• Design Patterns: Appropriate pattern usage, SOLID principles adherence
• Best Practices: Language idioms, error handling, resource management

For detailed analysis criteria and thresholds, see [review-workflow.md](references/review-workflow.md)

For quality metrics and thresholds, see [quality-metrics-reference.md](references/quality-metrics-reference.md)

Step 4: Document Findings

Structure the review report with:

• Executive summary with scores and top priorities
• Detailed findings with severity, location, description, and recommendations
• Metrics summary with current vs. target values
• Prioritized recommendations (P0-P3)
• Positive observations acknowledging good practices
• Technical debt summary with effort estimates

For complete report structure and output guidelines, see [review-report-format.md](references/review-report-format.md)

Quality Assurance

Use the checklist to ensure comprehensive reviews:

• Code organization and structure
• Naming conventions and clarity
• Complexity thresholds
• Error handling patterns
• Testing and documentation
• Security considerations
• Performance implications

For complete checklist, see [best-practices-checklist.md](references/best-practices-checklist.md)

Common Pitfalls

Avoid these common review mistakes:

• Focusing only on style issues instead of substantive problems
• Being overly critical without actionable suggestions
• Ignoring context and business constraints
• Overwhelming with too many issues at once
• Using vague terms without explanation
• Forgetting to acknowledge good practices

For detailed guidance, see [common-pitfalls-to-avoid.md](references/common-pitfalls-to-avoid.md)

Example Patterns

For reference when identifying critical issues in your review, see examples of common high-severity problems in [critical-issues.md](references/critical-issues.md)

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: DauQuangThanh
• Source: DauQuangThanh/hanoi-rainbow
• License: MIT

Install and usage instructions live in the source repository linked above.