Inspired by this reddit post I wrote a scraper to save all of his mum's mylot posts to PDFs, including images and comments.
Apologies in advance for the messy code, I wrote it in about half an hour. Feel free to PR fix.
Run in like so, supplying the username of the person your wish to archive.
node ./index.js ridingbet First is paginates all the articles to get a complete list. Once that's done, using 5 instances of puppeteer (headless browser) it creates PDF copies of each article. There's a little bit of trickery to get it looking good (hiding elements/etc).
The PDFs are named with the ISO timestamp first, so they sort nicely in the folder from oldest to newest
MIT, do you want want :)