AI Code Review Pro is a powerful, extensible CLI tool and GitHub Action that automates code review using OpenAI’s language models. It helps teams maintain code quality, enforce style guidelines, and catch potential bugs before merging.

• Modular Architecture: Core engine, CLI layer, plugin system and output handlers are decoupled.
• Multi-Language Support: JavaScript, TypeScript, Python, Go, Rust, Java (easily extendable).
• Plugin System: Write and share custom linting and style rules.
• Flexible Outputs: Terminal, JSON, HTML, Markdown report formats.
• Configurable Thresholds: Control sensitivity for style, security and best-practice suggestions.
• GitHub Action: Automatically review pull requests on opened, synchronize, or reopened events.
• Detailed Logs & Metrics: Structured logging via Winston, metrics for review duration and issue counts.
• Built-In Testing: Jest-based test suite to validate parser and plugin behavior.

• Node.js v16 or later
• npm v8 or later
• An OpenAI API Key (GPT-4 capable)
• GitHub repository with main branch

1. Clone the repository

   git clone https://github.com/icelaterdc/AI-Code-Review-Pro.git
   cd AI-Code-Review-Pro

2. Install dependencies

   npm install

3. Copy environment template

   cp .env.example .env

4. Set your OpenAI key in .env

   OPENAI_API_KEY=sk-...
   

1. Initialize default configuration

   npx AI-Code-Review-Pro init

2. Run a review on your source folder

   npx AI-Code-Review-Pro review src/ --output html,json

3. View report

   • Terminal output summarized
   • review-report.html autogenerated in project root

Usage: ai-review-pro [options] [command]
Commands:
 review <path> Review code at specified file or directory
 init Generate default `aireview.config.js`
 plugin:install <plugin>
 Install a plugin from npm or local path
 help [command] Display help for command
Options:
 -V, --version output the version number
 -h, --help display help for command

• Review single file

  npx AI-Code-Review-Pro review src/index.js

• Review with JSON output

  npx AI-Code-Review-Pro review src/ -o json

• Install community plugin

  npx AI-Code-Review-Pro plugin:install @my-org/review-plugin-security

Default config file: aireview.config.js in project root.

module.exports = {
 openaiModel: 'gpt-4o-mini', // Model name
 threshold: 0.75, // Confidence threshold (0.0 - 1.0)
 languages: ['js','ts','py'], // Extensions to review
 output: ['terminal','html'], // Output formats
 plugins: [ // Plugins to load
 './plugins/sample-plugin.js'
 ],
 rules: { // Built-in rule overrides
 'no-todo-comments': true,
 'max-line-length': 120
 }
};

• openaiModel: Change to any supported OpenAI model.
• threshold: Lower thresholds generate more suggestions.
• languages: File extensions to include.
• output: terminal, json, html, markdown.

Create a plugin by extending the base Plugin class:

import { Plugin } from 'ai-review-pro/src/plugins/base.js';
export default class MyCustomPlugin extends Plugin {
 apply(review) {
 const { text, issues } = review;
 if (/eval\(/.test(text)) {
 review.addIssue({
 message: 'Avoid using eval(), it may introduce security risks.',
 severity: 'error',
 });
 }
 }
}

• Place plugin file under plugins/
• Reference in plugins array of config
• Plugins can manipulate review.issues and metadata

• Terminal: Colorized summary
• JSON: Raw data for integrations
• HTML: Styled report with sections per file
• Markdown: For GitHub PR comments

All reports are saved in ./review-report.[ext] by default.

Using Winston for structured logs:

{
 level: 'info',
 transports: [new winston.transports.Console()]
}

Metrics collected:

• Review start/end timestamps
• Files processed count
• Total issues found

Run unit tests:

npm test

Includes parser and plugin behavior tests. Coverage reports generated under coverage/.

1. Fork the repo
2. Create feature branch: git checkout -b feature/my-feature
3. Commit changes: git commit -m 'Add new plugin support'
4. Push: git push origin feature/my-feature
5. Open a Pull Request

Please follow the Code of Conduct and Contributor Guidelines.

MIT © 2025 IceLater

Q: How do I exclude files or directories? A: Add a .aireviewignore file with glob patterns to exclude.

Q: Can I review only staged changes? A: Yes, use --staged flag in review command (coming soon).

Q: Which OpenAI models are supported? A: Any model available through your API key; default is gpt-4o-mini.