I built BlurrySite, an Chrome extension focused on packing the most needed privacy tools into one place; There are around 3 primary features / 5 small features / lot of customisations. I gave my best to have a lot of depth as a dev. I demoed Pick and Blur / Screen Share feature below, there is Sensitive info detection / Blur All. This is my first side project, so i dont know how to market what this extension could at the fullest, there is not one extension as Blurry Site in chrome. I am dev, i did what i could do best with UI; i still think UX sucks and has a learning curve, any ideas or feedback is appreciated

demo

>Its a privacy tool so there will be Zero network requests. No analytics. No telemetry. No tracking. No accounts. No payments. Everything stays in Chrome on your device. The extension does not even have a remote update channel beyond the browser's own.

A little bit of backstory -> A friend asked me, if I could build a tool that would blur content on shortcut to avoid over the shoulder peeks at workplaces. I thought that was needed and thus the first feature Blur All. Also, you can reveal applied blur on hover per element and a lot of customization underneath. Configure your own shortcuts.

There is also Pick & Blur option if you want to blur by drawing stuff on top of webpage. Later, i implemented Sensitive Info detection (automatically finds emails, card numbers, and other financial info as best it can.).

Screen Share -> Sometimes, All it takes is one wrong tab with some data during a screen share. Well, with this you can guard your privacy to avoid such situations.

works on all tabs / sites. Offers suppression if required.

Chrome Store Link -> https://chromewebstore.google.com/detail/blurry-site/nceghmchnpfippfofmbagckbinnkgaje

Github Link -> https://github.com/ChennakeshavaG/blurrysite

https://i.redd.it/n59e8q9wo30h1.gif

I built BlurrySite, an Chrome extension focused on packing the most needed privacy tools into one place; If you don't like my approach or UI, fork it and build your own, code is open sourced.

>Zero network requests. No analytics. No telemetry. No tracking. No accounts. No payments. Everything stays in Chrome on your device. The extension does not even have a remote update channel beyond the browser's own.

A little bit of backstory -> A friend asked me, if I could build a tool that would blur content on shortcut to avoid over the shoulder peeks at workplaces. I thought that was needed and thus the first feature Blur All. Also, you can reveal applied blur on hover per element. Configure your own shortcuts.

There is also Pick & Blur option if you want to blur by drawing stuff on top of webpage. Later, i implemented Sensitive Info detection (automatically finds emails, card numbers, other financial number), the approach is using a layers of regex, it will get job done for most use cases.

Screen Share -> Sometimes, All it takes is one wrong tab during a screen share. The extension automatically blurs other tabs on screen share and provides option to suspend per tab / site / temporarily for session until browser restarts. You can enable them through extension popup or toasts.

https://i.redd.it/ifilkvbyo30h1.gif

I also included blur on idle / tab switch. site rules if needed.

There is some options under customize for depth, do check them out!!!

The code is open sourced

Github repo -> https://github.com/ChennakeshavaG/blurrysite

Chrome Extension Link -> https://chromewebstore.google.com/detail/blurry-site/nceghmchnpfippfofmbagckbinnkgaje

I built it as hobby side project. Hope, you will find it useful.

Processing img ygxj3y8cj10h1...

I built BlurrySite, an Chrome extension focused on packing the most needed privacy tools into one place; If you don't like my approach or UI, fork it and build your own, code is open sourced.

>Zero network requests. No analytics. No telemetry. No tracking. No accounts. No payments. Everything stays in Chrome on your device. The extension does not even have a remote update channel beyond the browser's own.

A little bit of backstory -> A friend asked me, if I could build a tool that would blur content on shortcut to avoid over the shoulder peeks at workplaces. I thought that was needed and thus the first feature Blur All. Also, you can reveal applied blur on hover per element. Configure your own shortcuts.

There is also Pick & Blur option if you want to blur by drawing stuff on top of webpage. Later, i implemented Sensitive Info detection (automatically finds emails, card numbers, and other financial info as best it can.), the approach is using a layers of regex, not perfect per se, it will get job done in most use cases.

Screen Share -> Sometimes, All it takes is one wrong tab during a screen share.

this works across all screen sharing websites

And here we are, I gave small glimpse of what I did. Honestly probably went too far on features; let me know what's actually useful and what needs some tuning. Overall, its a fun side project...

Chrome Extension Link -> https://chromewebstore.google.com/detail/blurry-site/nceghmchnpfippfofmbagckbinnkgaje

Github repo -> https://github.com/ChennakeshavaG/blurrysite

So, dev stuff;

FYI -> My code will smell like AI generated, because 100% is. I built this over a span of a month vibe coding in my free time. I took up this personal project to learn how to solve problems and build real products without writing a single piece of code (a hard rule). Yep!! That was the primary intention to start this, I didn't write one single line of code. I used claude code cli for even changing versions lol, which is a 2 line change. Then there is cool UI / gifs / tooltips. I didnt lay hands on them either.

I learnt a lot on how to use LLMs effectively, they are really good at solving problems quickly. Before LLMs, jumping into an unfamiliar area of expertise and just building stuff out wasn't really possible. Now it is. I am primarily an backend dev; frontend / extension development is not my forte.

These are the hated things during the development; Iframes, Shadow Roots (wtf is wrong with reddit), 5+ Nested Div to show a text (looking at you whatsapp). I tried my best to solve it, there are still some improvements needed. And of course claude, it's either does work very smart or irritatingly dumb at times for the similar kind of work, its a hit or miss depending day and time.

Most of my time went to researching how things work / lot of iterations. I rewrote lot of pieces more than 5-6 times and redid my entire code architecture more than 3 times, had gone through 3 different UI before finalizing on current one. The bottleneck is no longer pushing code, rather researching / solving / building to getting it right. As I was using Claude more, I realized you can really go an extra mile and get done quickly which is not possible before LLMs.

Easier to burn your hands on side projects than on production systems. I did burn... codeburn says I burned around 2k dollars worth of tokens (I did burn a lot for sure), In reality I used Claude Max 5x to the fullest, i hit daily quotas, weekly quotas all the time in last 1 month. Absolutely worth it. Claude can be pain with unreliability at times.

It's been an interesting month. Firefox testing is still in progress. Might finish up, once I find time I guess. If you want anything yourself, you can fork the code, build yourself and run it locally. I will maintain the project or assist as long as my privilege and life permits.

If you're wondering how this screen share works or any other feature, feel free to ask in comments. I will explain my best on how i built it. Do try it out. The extension is still rough on the edges. If you find any bugs, do report in my github repo.

Processing img ygxj3y8cj10h1...