ChatGuard by Vigilnz - Community Edition

About

ChatGuard provides real-time security analysis for AI chat interfaces. It monitors prompts and responses to detect potential security threats like prompt injection, jailbreak attempts, system prompt extraction, and data exfiltration.

How it works:
- Analyzes text in real-time using OpenAI's API (requires your own API key)
- Displays safety badges on each message showing Safe/Unsafe status with confidence percentage
- Maintains a local detection log for review and CSV export
- Works across popular LLM chat platforms

Features:
- Real-time input monitoring and message scanning
- Visual safety indicators with color-coded badges
- Context-aware badge placement that adapts to different chat interfaces
- Local detection history with detailed logs
- CSV export functionality for analysis
- Customizable settings with auto-save
- Toggle to show/hide visual annotations

Privacy:
All analysis is performed using your own OpenAI API key. Detection logs are stored locally on your device and never transmitted to external servers. The extension only accesses chat interfaces you visit and does not collect personal information.

Setup:
1. Install the extension
2. Open the popup and navigate to Settings
3. Enter your OpenAI API key
4. Enable the extension
5. Visit any supported chat interface to begin monitoring

ChatGuard helps you stay aware of potential security risks in your AI conversations while maintaining full control over your data.