AI Comment Guard

🤖 AI-powered comment moderation for WordPress - Protect your WordPress site from spam with intelligent, customizable AI analysis.

🌟 Features

• 🔌 Multiple AI Providers: OpenAI (GPT-4/GPT-3.5), Anthropic (Claude), OpenRouter
• ⚡ Automatic Processing: Real-time comment analysis and moderation
• 🎯 Smart Classification: Auto-approve, reject, hold, or mark as spam
• 📊 Confidence Thresholds: Customizable confidence levels for actions
• ✏️ Custom Prompts: Tailor AI behavior to your needs
• 📈 Comprehensive Logging: Track all decisions with detailed statistics
• 🔒 Secure Integration: Encrypted API key storage
• 🌍 i18n Ready: Fully translatable
• ⚙️ Easy Setup: Intuitive admin interface

📋 Requirements

• WordPress 5.0+
• PHP 7.2+
• API key from supported provider
• SSL certificate (recommended)

🚀 Installation

From WordPress Admin

1. Navigate to Plugins > Add New
2. Search for "AI Comment Guard"
3. Click Install Now then Activate
4. Go to Settings > AI Comment Guard

Manual Installation

1. Download the latest release
2. Upload to /wp-content/plugins/ai-comment-guard/
3. Activate through the Plugins menu
4. Configure at Settings > AI Comment Guard

⚙️ Configuration

1. Choose Provider: Select OpenAI, Anthropic, or OpenRouter
2. Add API Key: Enter your provider's API key
3. Test Connection: Verify the connection works
4. Set Thresholds: Configure confidence levels
5. Customize Prompts: Tailor AI analysis criteria
6. Enable Logging: Track moderation decisions (optional)

🔑 API Providers

OpenAI

• Get your key at platform.openai.com
• Supports GPT-4 and GPT-3.5 models
• Excellent for nuanced content analysis

Anthropic

• Get your key at console.anthropic.com
• Uses Claude models
• Great for context-aware moderation

OpenRouter

• Get your key at openrouter.ai
• Access to multiple AI models
• Flexible pricing options

📊 Usage Examples

Basic Setup

// The plugin works automatically once configured
// Comments are processed before being saved to database

Custom Threshold Example

• Spam Threshold: 0.7 (70% confidence = mark as spam)
• Approval Threshold: 0.3 (30% confidence = auto-approve)
• Between thresholds: Hold for manual review

Custom Prompt Example

Analyze this comment for spam, inappropriate content, or legitimate discussion.
Consider: relevance, tone, promotional content, and value to discussion.
Respond with JSON: {"analysis": "approved|rejected|spam", "confidence": 0.0-1.0, "reason": "explanation"}

🛡️ Security Features

• Nonce Verification: All AJAX requests protected
• Capability Checks: Admin-only access to settings
• Data Sanitization: All inputs properly sanitized
• SQL Injection Protection: Prepared statements used
• XSS Prevention: Output properly escaped
• HTTPS Only: API communications encrypted

🌐 Internationalization

The plugin is fully translatable with .pot file included. Available text domains:

• ai-comment-guard

To translate:

1. Use the included .pot file
2. Create your .po and .mo files
3. Place in /wp-content/languages/plugins/

📈 Performance

• Async Processing: Non-blocking comment analysis
• Database Indexes: Optimized query performance
• Caching: Configuration cached for efficiency
• Cleanup Cron: Automatic old log removal
• Minimal Overhead: < 50ms average processing time

🔧 Development

File Structure

ai-comment-guard/
├── admin/              # Admin interface assets
│   ├── css/           # Admin styles
│   └── js/            # Admin scripts
├── includes/          # Core PHP classes
│   ├── AI/           # AI provider implementations
│   ├── Admin/        # Admin functionality
│   ├── Comments/     # Comment processing
│   ├── Core/         # Core plugin files
│   ├── Database/     # Database operations
│   └── Utils/        # Utility classes
├── languages/         # Translation files
├── ai-comment-guard.php  # Main plugin file
├── readme.txt         # WordPress.org readme
├── README.md          # This file
└── uninstall.php      # Clean uninstall handler

Coding Standards

• PSR-4 Autoloading: Modern PHP namespace structure
• WordPress Coding Standards: Following WP guidelines
• PHPDoc Comments: Complete documentation
• Security Best Practices: OWASP guidelines followed
• Design Patterns: Singleton, Factory, Strategy patterns

Hooks & Filters

// Filter comment approval status
add_filter('pre_comment_approved', 'your_function', 10, 2);

// Action after AI analysis
do_action('aicog_after_analysis', $comment_data, $analysis);

// Filter AI prompt
add_filter('aicog_prompt', 'customize_prompt', 10, 2);

🧪 Testing

Manual Testing Checklist

• Install and activate plugin
• Configure API provider
• Test connection
• Submit test comment
• Verify AI analysis
• Check logs (if enabled)
• Test threshold adjustments
• Verify uninstall cleanup

📝 Changelog

For detailed version history and release notes, see the changelog in readme.txt or visit the WordPress.org plugin page.

🤝 Contributing

Contributions are welcome! Please:

1. Fork the repository
2. Create a feature branch
3. Commit your changes
4. Push to the branch
5. Open a pull request

Reporting Issues

Please report issues with:

• WordPress version
• PHP version
• Error messages
• Steps to reproduce

📜 License

AI Comment Guard is licensed under GPL v2 or later.

This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version.

👨‍💻 Author

Tudor Constantin

• LinkedIn: tudor-eusebiu-constantin

🙏 Acknowledgments

• WordPress Community for feedback and support
• Contributors and testers
• AI provider teams for excellent APIs

📞 Support

For support, please:

1. Check the FAQ section
2. Visit the support forum
3. Contact via LinkedIn

---

⭐ If you find this plugin useful, please consider leaving a 5-star review on WordPress.org!