Klarity

Reputable Site Sample
Questionable Site Sample

Inspiration

Spread of misinformation in the media around us. Bad actors now weaponize presentation, using deceptive design, emotional manipulation, and fake virality. Studies show that nearly half of teens actively fall for fake news because engagement-driven platforms and technology has outpaced our traditional digital literacy education. We also noticed that current fact-checking tools act as passive censors, where they might flag a site as "false," but they fail to teach the user why.

What it does

Klarity is a real-time browser extension that acts as a transparent credibility dashboard to evaluate website reliability. It breaks down common patterns seen in bad sources by generating a Credibility Score based on four key metrics. The extension instantly analyzes a site's critical authenticity signals, editorial standards, consensus indicators, and page hygiene to provide a comprehensive breakdown of a site's trustworthiness.

How we built it

We built the frontend using HTML, CSS, and JavaScript, leveraging Chrome Extension APIs to create a clean, modern UI with visual meters and custom SVG branding that interacts directly with the user's active tab. To power our backend logic, we integrated the Google Gemini API as the core engine of the Klarity Credibility Algorithm. By utilizing Gemini for real-time semantic and structural analysis, we modeled information reliability as a bounded, multi-layered scoring function featuring an additive baseline-delta architecture.

Challenges we ran into

One of the biggest challenges was mathematically modeling "credibility" without building a biased censor. Formulating the baseline-delta architecture took careful iteration to ensure we were penalizing actual spam and content farms, rather than just websites with poor web design.

Accomplishments that we're proud of

We're very proud of our frontend execution. The UI is highly responsive, visually striking, and clearly communicates risk to the user without overwhelming them with a lot of data.

What we learned

We learned about some common patterns present in pages with misinformation. On the technical side, we deepened our understanding of Chrome Extension architecture, and how to effectively build user-facing interfaces that process background data in real-time.

What's next for Klarity

Our next major step is to expand the "Inspect Highlights" feature, injecting JavaScript directly into the page to visually highlight the exact hyperbolic trigger words and un-cited claims in the text. This will further guide users on patterns present on misinformation pages/sources.