yelhouti

LinkedIn scans your browser for 4,934 Chrome extensions. Here is how to see it yourself.

I was debugging an unrelated console error on linkedin.com last week and noticed something odd: hundreds of failed requests to chrome-extension://invalid/.

Here is what is going on.

Any web page can test whether you have a specific Chrome extension installed by requesting a file from it. If the extension is there, the file loads. If it is not, Chrome rewrites the URL to chrome-extension://invalid/ and the request fails silently. No permission needed, nothing visible to the user.

LinkedIn's page bundle walks a hardcoded list of extension IDs and probes each one in turn. I logged the list. It contains 4,934 entries, in alphabetical order, each paired with a real file path inside that specific extension: icon16.png, content.css, assets/index-DgZXh0UX.js. Nobody writes those by hand. Someone crawled the Chrome Web Store, downloaded the extensions, parsed their manifests and built a detection list.

yelhouti

# LinkedIn scans your browser for 4,934 Chrome extensions. Here is how to see it yourself.

I was debugging an unrelated console error on linkedin.com last week and noticed something odd: hundreds of failed requests to `chrome-extension://invalid/`.

Here is what is going on.

Any web page can test whether you have a specific Chrome extension installed by requesting a file from it. If the extension is there, the file loads. If it is not, Chrome rewrites the URL to `chrome-extension://invalid/` and the request fails silently. No permission needed, nothing visible to the user.

LinkedIn's page bundle walks a hardcoded list of extension IDs and probes each one in turn. I logged the list. It contains 4,934 entries, in alphabetical order, each paired with a real file path inside that specific extension: `icon16.png`, `content.css`, `assets/index-DgZXh0UX.js`. Nobody writes those by hand. Someone crawled the Chrome Web Store, downloaded the extensions, parsed their manifests and built a detection list.

yelhouti

Code Generation

There two main approches of handling upgrades when using a code generator:

1. The generated code is never touched (only extended...), and regeneration only updates these files
2. Generated code is stored in a separate branch where generation is done no changes are done manually, the generated branch is merged with master at each upgrade

I find the first approach hard to work with for two reasons:

1. some files can't be extended like package.json ...
2. Extending all the files makes for bigger code base and adding simple changes can add a lot of work

yelhouti

   0.0 TEL | Telepresence 0.99 launched at Mon Jun  3 21:21:08 2019
   0.0 TEL |   /usr/bin/telepresence --docker-run --rm dbmigration:latest
   0.0 TEL | Platform: linux
   0.0 TEL | Python 3.6.8 (default, Jan 31 2019, 09:38:34)
   0.0 TEL | [GCC 8.2.1 20181215 (Red Hat 8.2.1-6)]
   0.0 TEL | [1] Running: uname -a
   0.0   1 | Linux yelhouti-p52 4.20.14-100.fc28.x86_64 #1 SMP Tue Mar 5 19:45:42 UTC 2019 x86_64 x86_64 x86_64 GNU/Linux
   0.0 TEL | [1] ran in 0.00 secs.
   0.0 TEL | BEGIN SPAN main.py:40(main)
   0.0 TEL | BEGIN SPAN startup.py:74(__init__)

yelhouti

<application>
    <navigation-handler>
        org.primefaces.mobile.application.MobileNavigationHandler
    </navigation-handler>
    <default-render-kit-id>PRIMEFACES_MOBILE</default-render-kit-id>
</application>

<render-kit>
<render-kit-id>PRIMEFACES_MOBILE</render-kit-id>
    <renderer>