its mostly because MCPs handle auth in a standardised way and give you a framework you can layer things like auth, etc on top of.

Without it youre stuck with the basic http firewall, etc which is extremely dangerous and this is maybe the 1 opportunity we have to do this.

Both scale and purpose make them completely different things. You're acting as if they're the same when they're not.

I won't comment about dl but ads are trackers and spyware for me. I don't spy on websites' owners, I have my human rights to stop those trackers.

Zuck serves ads/spywares to other users, he deserves to taste his own medicines, not me.

You can see this pattern in many different topics: updoots are highly correlated with a positive answer to "do I personally get to profit"?

I would love to pay for content. I'm _paying_ for YouTube Premium.

But heck. Do I hate the YouTube interface, it degraded far past usability.

So you’re that Hal Jordan then? Why would a Green Lantern feel the need to defend either? I feel like the Guardians would not accept your arguments as soon as you got to Oa, poozer. I guess what I am saying is don’t have a famous name. Seems obvious.

you mean this one? https://github.com/vercel-labs/agent-browser

You can just start claude with the —chrome flag too and it will connect to the chrome extension.

have you tried vibium's cli + agent skill?

Don’t know how this could be more stable than ytdlp. When issues come up they’re fixed really quickly.

> yt-dlp arms race

I don't know anything about yt-dlp.

It would probably help people who want to go to a concert and have a chance to beat the scalpers cornering the market on an event in 30 seconds hitting the marketplace services with 20,000 requests.

I can try to see if can bypass yt-dlp. But that is always a cat and mouse game.

If it can save all the video/audio fragment and call ffmpeg to join them together. Maybe?

Exactly, it is an agent skill that interacts pressing buttons and stuff with a webpage capturing and documenting all the API requests the page makes using Playwright's request / response interception methods. It creates and strongly typed well documented API at the end.

It turns any authenticated browser session into a fully typed REST API proxy — exposing discovered endpoints as local Hono routes that relay requests through the browser, so cookies and auth are automatic.

The point is that it creates an API proxy in code that a Typescript server calls directly. The AI runs for about 10 minutes with codegen. The rest of the time it is just API calls to a service. Remove the endpoint for "Delete Account" and that API endpoint never gets called.

This 100% breaks everyone's terms of service. I would not recommend nor encourage using.

100% I'll response to this by Friday with link to Github.

I use Patchright + Ghostery and I have a cleaver tool that uses web sockets to pass 1 second interval screenshots to the a dashboard and pointer / keyboard events to the server which allow interacting with websites so that a user can create authentication that is stored in the chrome user profile with all the cookies, history, local storage, ect.. in the cloud on a server.

Can you list some websites that don't require subscription that you would like to me to test against? I used this for Robinhood and I think Linked in would be a good example for people to use.