MorisGuard

Detect toxicity in Mauritian Creole comments using an AI-powered moderation pipeline and Machine Learning.

MorisGuard is a specialized moderation extension designed to detect toxicity and hate speech in Mauritian Creole text. Leveraging a sophisticated three-layer AI pipeline, MorisGuard provides real-time insights into the sentiment and safety of online comments, ensuring that linguistic nuances are never lost in translation. Key Features: Multi-Layered Analysis: Every text is processed through a specialized dictionary, a custom Machine Learning model, and advanced Generative AI (Gemini) for maximum accuracy. Instant Context Menu Integration: Simply highlight any text on a webpage, right-click, and select "Detect Toxicity with MorisGuard" to get an immediate safety report. Seamless Side Panel Experience: Open the side panel to manually paste and analyze text without leaving your current tab or interrupting your workflow. Visual Safety Indicators: Get instant feedback with color-coded results—Safe (Green), Human Review (Orange), or Unsafe (Red)—making moderation tasks faster and more intuitive. Deep-Dive Analysis: One-click access to the MorisGuard web dashboard for a comprehensive breakdown of findings. Why MorisGuard? General-purpose moderation tools often struggle with the unique syntax and slang of Mauritian Creole. MorisGuard is built specifically for the Mauritian digital landscape, making it an essential tool for content creators, community managers, and researchers working with local content. Privacy & Safety: MorisGuard is designed with privacy in mind. We only analyze the text you explicitly submit or select, and our focus is entirely on identifying harmful content to create a safer internet for everyone.