Remix clone Hacker News

new | show | ask | jobs Github

▲

echelon 3 hours ago

> where I had an agent karma

Was this a browser using agent? What did you use?

▲

carlgreene 3 hours ago | parent [-]

It used the browser agent to grab user cookies after signing in, then made API calls iirc.

Using just a browser is way too token intensive and slow. It would look for 401 errors then run the browser automation to login with the credentials and grab the token.

▲

echelon 3 hours ago | parent [-]

I'm surprised these platforms don't have advanced heuristics to detect API calls and inauthentic traffic.

Did you clone the Reddit API from browser traffic and then turn it into a 100% API driven thing?

I'd imagine they'd be sniffing browser agents, plugins, cookies, etc. to fingerprint. Using JavaScript scroll position, browsing rate and patterns, etc.

Maybe their protections just aren't that sophisticated.

▲

tardedmeme 2 hours ago | parent [-]

Reddit is known to fingerprint TLS and quickly shadowban accounts that don't have the fingerprints of browsers.

	▲	echelon 2 hours ago \| parent [-]
		TLS fingerprinting and Cloudflare are easy to bypass. There are lots of libraries that do so. The application-layer stuff is harder. Each application can develop its own heuristics, and that's difficult to automate in a cross-cutting fashion.