It ought to be trivially possible, going on from things like grease monkey and ad blocking scripts to add a bit of user intent guessing / content filtering / simplified view logic.
But there's always an arms race with stuff like this. Our robot butler would be able to figure out how to parse the web today, but the advertisers would completely change the structure of pages tomorrow.
Facebook used to have their ads wrapped in `<div class='advertisement'>`. Now, they're buried a thousand divs deeps, and the ads and content both have nonsensical, non-semantic class names.
People in the know have the right extensions to block stuff and get past paywalls using the right snapshot services etc. Firefox can't officially offer this, but they can let people do whatever they want if they run their own extensions.
If this was common place or convenient for the masses then it would be actively discouraged or outlawed, similar to what happened with Napster and MP3s.
Sorry but Firefox is mostly on the dark side these days. That shouldn't be too surprising since most of its money comes from google.
It does have a setting to stop embedded video from playing sound unless you send an event, but it is too easy for obnoxious pages to evade and play sound against your wishes.
> I have a dream for the Web [in which computers] become capable of analyzing all the data on the Web – the content, links, and transactions between people and computers. A "Semantic Web", which makes this possible, has yet to emerge, but when it does, the day-to-day mechanisms of trade, bureaucracy and our daily lives will be handled by machines talking to machines. The "intelligent agents" people have touted for ages will finally materialize.
- Tim Berners Lee
So you'd need a team, and as somebody with a tech and a library background, good luck getting those two groups to communicate and work together. Doing it without the information seeking background just recreates the issues with the modern web where your 'robot butler' comes back with a ton of irrelevant information, and doing it without the tech background just gets you a bunch of librarians high-fiving themselves over 'solving' a problem using outdated tech and is used by 20 people because it doesn't scale well.
Unfortunately we aren't very good at making robot butlers yet, and copyright severely limits what they can do (provided they are running as a third party service) in the first place.
That's what an RSS reader does.