AI Code Reviewer

An intelligent, AI-powered code review tool that integrates seamlessly with GitHub and supports multiple AI providers (OpenAI, Anthropic). Perfect for organizations looking to maintain code quality standards across multiple repositories.

🚀 Features

• Multi-AI Provider Support: Works with OpenAI GPT-4o and Anthropic Claude
• GitHub Integration: Automatic PR comments, approvals, and status checks
• Configurable Standards: Custom coding standards and review prompts
• CLI & GitHub Actions: Local usage and CI/CD integration
• Smart Filtering: Ignores irrelevant files and focuses on meaningful changes
• Severity Levels: Categorizes issues by severity and type
• Organization Ready: Reusable configurations across multiple repositories

📦 Installation

npm install -g code-review-agent

🎯 Quick Start

1. Initialize Configuration

ai-review init

This creates:

• .ai-review.json - Main configuration
• coding-standards.md - Your coding standards
• review-prompt.md - AI review instructions

2. Set API Keys

export OPENAI_API_KEY="your-openai-key"
# OR
export ANTHROPIC_API_KEY="your-anthropic-key"

3. Review Code Changes

# Review current changes
ai-review review

# Review specific commits
ai-review review --base main --head feature-branch

🔧 Configuration

Basic Configuration (.ai-review.json)

{
  "aiProvider": "openai",
  "model": "gpt-4o",
  "maxTokens": 2000,
  "temperature": 0.3,
  "codingStandardsPath": "coding-standards.md",
  "reviewPromptPath": "review-prompt.md",
  "ignorePatterns": [
    "node_modules/**",
    "dist/**",
    "*.min.js"
  ],
  "severityThreshold": "medium",
  "autoApprove": false,
  "requireApprovalFor": ["error"]
}

Advanced Configuration Options

| Option | Description | Default | |--------|-------------|---------| | aiProvider | AI provider (openai or anthropic) | "openai" | | model | AI model to use | "gpt-4o" | | maxTokens | Maximum tokens per request | 2000 | | temperature | AI creativity (0-1) | 0.3 | | codingStandardsPath | Path to coding standards file | "coding-standards.md" | | reviewPromptPath | Path to review prompt file | "review-prompt.md" | | ignorePatterns | Files/patterns to ignore | [] | | severityThreshold | Minimum severity to report | "medium" | | autoApprove | Auto-approve if no issues | false | | requireApprovalFor | Severities that block approval | ["error"] |

🔗 GitHub Integration

GitHub Actions Setup

1. Add the workflow file .github/workflows/ai-review.yml:

name: AI Code Review

on:
  pull_request:
    types: [opened, synchronize, reopened]

jobs:
  ai-code-review:
    runs-on: ubuntu-latest
    permissions:
      contents: read
      pull-requests: write
      checks: write
      
    steps:
      - name: Checkout code
        uses: actions/checkout@v4
        with:
          fetch-depth: 0
          
      - name: Setup Node.js
        uses: actions/setup-node@v4
        with:
          node-version: '18'
          
      - name: Install AI Code Reviewer
        run: npm install -g code-review-agent
        
      - name: Run AI Code Review
        env:
          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
          OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
        run: |
          ai-review github-action \
            --owner ${{ github.repository_owner }} \
            --repo ${{ github.event.repository.name }} \
            --pr ${{ github.event.pull_request.number }}

2. Add your AI provider API key to repository secrets:
   • OPENAI_API_KEY for OpenAI
   • ANTHROPIC_API_KEY for Anthropic

Manual GitHub Integration

export GITHUB_TOKEN="your-github-token"
ai-review github-action --owner myorg --repo myrepo --pr 123

📋 Coding Standards

Create a coding-standards.md file to define your project's standards:

# Coding Standards

## General Guidelines
- Write clean, readable, and maintainable code
- Follow consistent naming conventions
- Add meaningful comments and documentation

## Security
- Validate all inputs
- Use parameterized queries
- Never commit secrets

## Performance
- Optimize for readability first
- Profile before optimizing
- Use appropriate data structures

🎯 Review Categories

The AI reviewer categorizes issues into:

• 🏗️ Code Quality: Structure, patterns, best practices
• 🔒 Security: Vulnerabilities, input validation, data exposure
• ⚡ Performance: Efficiency, scalability, resource usage
• 🧹 Maintainability: Complexity, error handling, documentation
• 🎨 Style: Formatting, naming conventions, consistency
• 🧪 Testing: Coverage, test quality, edge cases
• 📚 Documentation: Comments, README, API docs

💻 CLI Commands

ai-review review

Review code changes locally.

ai-review review [options]

Options:
  -b, --base <sha>     Base commit SHA
  -h, --head <sha>     Head commit SHA
  -c, --config <path>  Path to config file
  --no-github          Skip GitHub integration

ai-review init

Initialize configuration files.

ai-review init [options]

Options:
  --config-only  Only create config file

ai-review github-action

Run in GitHub Actions mode.

ai-review github-action [options]

Options:
  --owner <owner>      Repository owner (required)
  --repo <repo>        Repository name (required)
  --pr <number>        Pull request number (required)
  -c, --config <path>  Path to config file

🏢 Organization Setup

1. Create Organization Config Repository

my-org-standards/
├── .ai-review.json
├── coding-standards.md
├── review-prompt.md
└── README.md

2. Reference in Projects

{
  "aiProvider": "openai",
  "codingStandardsPath": "../my-org-standards/coding-standards.md",
  "reviewPromptPath": "../my-org-standards/review-prompt.md"
}

3. Shared GitHub Action

Create a reusable workflow in .github/workflows/ai-review.yml:

name: Reusable AI Review
on:
  workflow_call:
    secrets:
      OPENAI_API_KEY:
        required: true

jobs:
  review:
    uses: my-org/standards/.github/workflows/ai-review.yml@main
    secrets:
      OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}

🔧 Programmatic Usage

import { ReviewEngine, ConfigLoader, GitHubClient } from 'code-review-agent';

// Load configuration
const config = await ConfigLoader.load();

// Review changes
const reviewEngine = new ReviewEngine(config);
const result = await reviewEngine.reviewChanges('main', 'feature-branch');

// Post to GitHub
const github = new GitHubClient(process.env.GITHUB_TOKEN);
await github.addReviewComments('owner', 'repo', 123, 'sha', result.comments);

🎨 Custom AI Providers

Extend the tool with your own AI provider:

import { BaseAIProvider } from 'code-review-agent';

class CustomProvider extends BaseAIProvider {
  name = 'Custom';
  
  async review(code: string, prompt: string, config: ReviewConfig): Promise<ReviewComment[]> {
    // Your implementation
  }
}

🐛 Troubleshooting

Common Issues

API Key Not Found

Error: OPENAI_API_KEY environment variable is required

Solution: Set your API key in environment variables.

Git Repository Not Found

Error: Failed to get git changes

Solution: Ensure you're in a git repository with proper permissions.

GitHub Token Permissions

Error: Failed to add review comments

Solution: Ensure your GitHub token has pull_requests: write permission.

Debug Mode

Enable verbose logging:

DEBUG=ai-review:* ai-review review

📄 License

MIT

🤝 Contributing

1. Fork the repository
2. Create your feature branch
3. Commit your changes
4. Push to the branch
5. Create a Pull Request

📞 Support

• GitHub Issues
• Documentation