AI Code Review

Overview

QODRYX's AI Code Review automatically analyzes every pull request, providing instant feedback on:

• Code quality and maintainability
• Security vulnerabilities and risks
• Performance optimizations
• Best practices and coding standards
• Potential bugs and edge cases

Key Features

How It Works

Setup

GitHub Integration

Enable AI code review for your GitHub repositories:

1. Go to Settings → Integrations in your QODRYX dashboard
2. Click Connect GitHub and authorize the app
3. Select the repositories you want to enable
4. Configure review settings (see below)

GitLab Integration

Enable AI code review for your GitLab repositories:

1. Go to Settings → Integrations in your QODRYX dashboard
2. Click Connect GitLab and provide your access token
3. Select the projects you want to enable
4. Configure review settings

Configuration

Customize code review behavior with a qodryx.yml file in your repository:

# qodryx.yml
code_review:
  enabled: true
  
  # What to analyze
  checks:
    security: true
    performance: true
    best_practices: true
    bugs: true
    style: true
  
  # Severity threshold for blocking PRs
  block_on: critical  # critical, high, medium, low, or none
  
  # Files to ignore
  ignore:
    - "*.generated.ts"
    - "vendor/**"
    - "*.min.js"
  
  # Custom rules
  rules:
    - name: "no-console-log"
      pattern: "console.log"
      message: "Remove console.log before merging"
      severity: warning
    
    - name: "require-error-handling"
      pattern: "await.*fetch"
      require: "try.*catch|.catch"
      message: "API calls should have error handling"
      severity: high
  
  # Style guide reference
  style_guide: "./docs/STYLE_GUIDE.md"
  
  # Auto-approve low-risk changes
  auto_approve:
    enabled: true
    conditions:
      - files_changed < 5
      - no_critical_issues
      - only_docs_or_tests

Review Categories

🔒 Security

• SQL injection vulnerabilities
• Cross-site scripting (XSS)
• Hardcoded secrets and credentials
• Insecure dependencies
• Authentication/authorization issues

⚡ Performance

• N+1 query detection
• Memory leaks
• Inefficient algorithms
• Missing indexes
• Unnecessary re-renders (React)

🐛 Bug Detection

• Null pointer exceptions
• Off-by-one errors
• Race conditions
• Unhandled edge cases
• Type mismatches

📝 Code Quality

• Code duplication
• Complex functions needing refactoring
• Missing documentation
• Inconsistent naming
• Dead code

Example Review

Here's what an AI code review looks like in practice:

// Suggested fix:
const result = await db.query(
  'SELECT * FROM users WHERE id = $1',
  [userId]  // Use parameterized query
);

PR Status Checks

QODRYX can block or warn on PRs based on review findings:

Next Steps