KernPDF
KernPDF for IT teams
In short: KernPDF is a free web tool that turns a web article into a clean PDF. It's a static website: everything happens in the visitor's own browser, there are no accounts, uploads, cookies or advertising, and the article never leaves their device. To use it, allow kernpdf.com, plus fonts.googleapis.com, fonts.gstatic.com and cdn.jsdelivr.net for now (see below).
Addresses to allow
| Address | What it's for | Needed? |
|---|---|---|
kernpdf.com and www.kernpdf.com | The tool itself: its pages, scripts, fonts and the how-to animations | Yes |
img.kernpdf.com | The picture helper (below). Lets pictures from pasted or fetched articles go into the PDF | Recommended. Without it, KernPDF still works; some pictures may be left out |
fonts.googleapis.com and fonts.gstatic.com | Google Fonts: the page's typefaces | Yes, for now. KernPDF will soon serve this itself |
cdn.jsdelivr.net | jsDelivr: Mozilla's Readability, which finds the article on a page | Yes, for now. KernPDF will soon serve this itself |
static.cloudflareinsights.com | Cloudflare Web Analytics: a visitor count with no cookies and no personal data | No. Blocking it changes nothing for the user |
Everything is served over HTTPS only (with HSTS), from Cloudflare.
A suggested category
If your filter asks for one, KernPDF fits Information Technology or Computers and Internet (also described as productivity, business or document tools). It is not a proxy, an anonymiser or a file-sharing service.
What KernPDF does
- A person brings in an article: a page saved from their browser, copied and pasted, or a PDF. Their browser reads it, finds the article and leaves out the ads and menus.
- They choose what to keep and a style, and their browser makes the PDF (or a one-page brief, or a PowerPoint slide) and saves it to their device.
- Nothing about the article, the PDF or what the person does is sent to any server. There's no sign-in and nothing to install.
- Settings and a personal library of saved items are kept in the browser's own storage on that device.
The picture helper, img.kernpdf.com
Most websites don't let one page read another site's pictures, so a pasted article's pictures can't go into the PDF directly. KernPDF asks its picture helper to fetch each picture from the website that hosts it and hand it back. It is deliberately narrow:
- it only answers requests from KernPDF's own pages;
- it only returns picture files (JPEG, PNG, GIF, WebP, AVIF), never web pages, scripts or other files, and nothing over 15 MB;
- it refuses private, local and internal addresses;
- it limits how many pictures one address can ask for each minute;
- it keeps no record of what was fetched beyond a short cache, and tells search engines not to index it.
Only the picture's web address and the article's web address are sent to it. People can also turn it off in KernPDF, and pictures inside a page saved from the browser never use it.
Security
- A strict Content Security Policy,
X-Content-Type-Options: nosniff,frame-ancestors 'none'and a limited Permissions Policy (no camera, microphone, location or payment). - No third-party scripts, advertising or trackers (its fonts and article finder currently come from Google Fonts and jsDelivr; they'll soon be served from kernpdf.com).
- Fixed, versioned files: each release's scripts have their own file names, so nothing changes without a new release.
Contact
KernPDF is made and run by Jon Coleman. For allow-listing or security questions, email support@kernpdf.com. See also the Privacy Notice and Terms of Use.