Comment by andrethegiant

Comment by andrethegiant 5 days ago

21 replies

View on Hacker News

I'm working on pure.md[1], which lets your scripts, APIs, apps, agents, etc reliably access web content in markdown format. Simply prefix any URL with `pure.md/` and you get the unblocked markdown content of that webpage. It avoids bot detection and renders JavaScript-heavy websites, and can convert HTML, PDFs, images, and more into pure markdown.

pure.md acts as a global caching layer between LLMs and web content. I like to think of it like a CDN for LLMs, similar to how Cloudinary is a CDN for images.

[1] https://pure.md

WillAdams 4 days ago

It seems to miss URLs?

At: https://willadams.gitbook.io/design-into-3d/2d-drawing the links for:

https://mathcs.clarku.edu/~djoyce/java/elements/elements.htm...

https://mathcs.clarku.edu/~djoyce/java/elements/bookI/bookI....

https://mathcs.clarku.edu/~djoyce/java/elements/bookI/defI1....

are rendered as:

_Elements_ _:_ _Book I_ _:_ _Definition 1_

Maybe detect when a page is on gitbook or some other site where there is .md source on github or some other site and grab the original instead?

Reply View 1 reply
  • andrethegiant 4 days ago

    By default, href values of <a> tags are removed, because they add significant token length without adding more context. Coming soon, you can specify a request header to set whether or not you want links removed from the response. Those underscores you mentioned are from the italics.

    Reply View 0 replies
wild_egg 5 days ago

Thanks for sharing. I was planning on building something like this in April after hitting too many issues with Jina and Tavily but it looks like you've already done the hard work!

Reply View 1 reply
wanderingbit 5 days ago

What a great idea, I will soon be a paying customer. This solves a problem of an app I'm using that I was hesitant to try to develop myself.

Reply View 1 reply
hardlyfun 5 days ago

Very nice, how did you manage to bypass sites with cloudflare turnstile setup?

Reply View 1 reply
erekp 4 days ago

how do you exactly fallback to common crawl? isn't the cost to even hold and query common crawl insane?

Reply View 2 replies
  • andrethegiant 4 days ago

    With AWS Athena, you can query the contents of someone else’s public S3 bucket. You pay per read, but if you craft your query the right way then it’s very inexpensive. Each query I run only scans about 1MB of data.

    Reply View 0 replies
sharpshadow 3 days ago

Works great on mobile thanks, helpful tool to bypass flaky websites, js and even some paywalls.

Reply View 0 replies
udev4096 5 days ago

[flagged]

Reply View 3 replies
  • NationOfJoe 5 days ago

    i have no skin in the game and honestly i am wondering how this idea contributes to enshittifying the web more?

    this idea just seems like it provides the same content as visiting the site in a different view, like reader mode?

    Reply View 2 replies
    • hbsbsbsndk 4 days ago

      The service seems designed to bypass anti-scraping measures. If site owners don't want their content scraped by AI this is subverting their will.

      It also obfuscates responsibility between the AI vendor and the scraping service. One can imagine unethical AI providers using a series of ephemeral "gateways" to access content while avoiding any legal or reputational harm.

      Reply View 0 replies
    • elric 4 days ago

      I think the parent is referring to the goal of making the web more "LLM friendly".

      Reply View 0 replies