code-review-ai

AI-powered architectural review and code quality analysis.

Overview

Why Recommended?

This plugin integrates with our development workflow. The global CLAUDE.md references code-review-ai:architect-review as the agent for code reviews before committing significant changes.

┌─────────────────────────────────────────────────────────┐
│  1. SPECIFY                                             │
│     - Define user story / task in CLAUDE.md             │
└─────────────────────────────────────────────────────────┘
                          │
                          ▼
┌─────────────────────────────────────────────────────────┐
│  2. IMPLEMENT                                           │
│     - Write code                                        │
│     - Write tests                                       │
└─────────────────────────────────────────────────────────┘
                          │
                          ▼
┌─────────────────────────────────────────────────────────┐
│  3. CODE REVIEW  ◄──── code-review-ai                   │
│     - Agent: code-review-ai:architect-review            │
│     - Incorporate feedback                              │
└─────────────────────────────────────────────────────────┘

Capabilities

Architectural Review

• Pattern compliance analysis
• Design principle adherence
• Dependency analysis
• Scalability assessment

Security Assessment

• Vulnerability detection
• Security best practices
• Authentication/authorization review
• Input validation checks

Code Quality

• Code smell detection
• Complexity analysis
• Maintainability scoring
• Test coverage assessment

Best Practices

• Language-specific conventions
• Framework-specific patterns
• Performance recommendations
• Documentation quality

Installation

Via Installer

./install.sh        # Fresh install
./install.sh --add  # Add to existing

code-review-ai is pre-selected by default.

Manual

# Add marketplace
claude plugin marketplace add wshobson/agents
 
# Install plugin
claude plugin install code-review-ai@claude-code-workflows

Usage

Direct Request

"Review this PR for architectural issues"
"Analyze the security of this module"
"Check this code for best practices"

Via Task Tool

The plugin provides a subagent that can be spawned via the Task tool:

subagent_type: "code-review-ai:architect-review"

This is used automatically when following the development workflow in global CLAUDE.md.

Review Output

The agent provides structured feedback:

1. Summary - Overall assessment
2. Architectural concerns - Design issues
3. Security findings - Vulnerabilities
4. Code quality - Improvements
5. Recommendations - Actionable next steps

Integration with Workflow

After implementing a feature:

1. Claude spawns code-review-ai:architect-review
2. Agent reviews the changes
3. Feedback is incorporated
4. Code is committed

This ensures consistent quality before commits.