ResiliSense

• 

  Home Page

• 

  Risk radar after analyzing a website

• 

  Action plan for user to better protect their identities

Inspiration

We live in a world where websites quietly profile us — from ad trackers to hidden scripts collecting personal data. We wanted to make that invisible process visible, so people can see how their privacy is being mirrored back to them.

What it does

ResiliSense scans any website, detects trackers, data-collection forms, and sensitive keywords, then generates a privacy “persona” — a reflection of what the site might infer about you. It grades the site’s privacy risk and visualizes it with an interactive radar chart.

How we built it

We built a full-stack system using Next.js for the frontend and Express + Playwright for the backend. The backend fetches or renders pages, analyzes scripts with Cheerio, and computes privacy risk scores. Then we use Google Gemini API and Google AI Studio to generate personas and contextual recommendations for safer browsing.

Challenges we ran into

Fetching and analyzing websites were difficult. Not only getting the context from the website is hard but also the fact that building a highly-precise heuristic model was challeging. Balancing performance with Playwright rendering and making the radar chart dynamically responsive were also tricky parts.

Accomplishments that we're proud of

Within 24 hours, I tried my best to implement a heuristic scoring model that is intentionally conservative: ad trackers, PII fields and sensitive-topic keywords are heavily penalized, and if a site combines ad-tracking with PII collection, we never grade it as A or B. It’s not a perfect academic model, but it’s transparent and explainable. I also managed to include AI Agent to produce a persona profile when being provided with context of the website

What we learned

We learned how privacy tracking operates under the surface, how to parse modern dynamic web pages, and how AI can turn raw data into meaningful, human-readable insights.

What's next for ResiliSense

We plan to expand detection accuracy, integrate real-time browser extensions, and let users compare sites or export privacy reports — moving from awareness to actionable privacy intelligence.

Built With

• express.js
• google-ai-studio
• next.js
• tailwind
• typescript