Raven

For more about our project take a look at our LinkedIn post (wink wink, stanley)!

https://www.linkedin.com/posts/nabil-aldihni_hackcanada2026-nixos-neovim-activity-7435920825972658176-yoDt?utm_source=share&utm_medium=member_desktop&rcm=ACoAACn5g5gBhwNfl9-rmj6njNtRwBWrnA6ChU8

Inspiration

We built Raven because “agreeing” to Privacy Policies and Terms of Service has become a required step for daily life, but actually reading them is unrealistic. Most people only realize the implications after something goes wrong: unexpected data sharing, aggressive tracking, or accounts that are hard to delete. We wanted a tool that makes those documents understandable in minutes, and keeping track of what’s collected, who it’s shared with, how long it’s kept, and what control the user has.

What it does

Raven is a privacy dashboard and integrated chrome extension focused on handing back control over your privacy and your data. In the dashboard, users are able to take a look into all the services affiliated with their accounts, and get a no nonsense interpretation of all those terms of service and privacy policies that no one ever bothers reading (but should!). The dashboard also gives you a little 'privacy score' based on the ways you say you care about the use of your data. As an awesome companion, the chrome extension lets users continue keeping track of all the

How we built it

Raven is a full‑stack web app built with a React + TypeScript frontend and a FastAPI (Python) backend, packaged to run locally via Docker Compose with a Postgres database.

Challenges we ran into

OAuth testing friction: Google OAuth apps in “testing” mode only work for approved test users, which blocked sign-in until we configured the consent screen correctly.
Permission vs functionality tradeoffs: We tried to keep Gmail access minimal, but discovered that ultra-restricted scopes can prevent common operations (like searching), forcing careful scope selection.
Performance + repeatability: Summarization can be slow/expensive, so we had to persist results and design the API so reloading a dashboard doesn’t re-run the whole pipeline.

Accomplishments that we're proud of

End-to-end product: Shipped a working full-stack app (React + FastAPI + Postgres) that runs locally via Docker Compose.
Consistent policy summaries: Turned long, messy Privacy Policies/ToS into structured, question-by-question summaries users can actually act on.