Go to file
Nick Thomas 3cc2c67949 Make it work on nodejs 13 2020-02-20 10:09:11 +00:00
.gitignore Initial commit 2017-01-26 18:00:36 +01:00
README.md Make it work on nodejs 13 2020-02-20 10:09:11 +00:00
main.js Make it work on nodejs 13 2020-02-20 10:09:11 +00:00
package.json Make it work on nodejs 13 2020-02-20 10:09:11 +00:00



This is a node script extracting E-Books from Amazon Kindle Cloud Reader. Useful e.g. for being able to read on devices where the Cloud Reader doesn't work and for having a copy in an open format. Works with Chrome/Chromium. Other browsers use different formats for the WebSQL file where the E-Books are stored for offline use.


  1. https://gist.github.com/yangchenyun/a1c123935d82f5e25d57
  2. https://gist.github.com/moodule/84140e557065ac3d73f669f120429ae1
  3. https://github.com/d10r/kindle-fetch

d10r added minor usability improvements (e.g. cmdline switch for input file selection) and a package.json file.

I have just updated it to cope with newer versions of Node.


This was successfully tested with Node v13.

npm install
./main.js --help

This generates HTML files with the book content in the directory you're in.

You just need to figure out the file to which the ebook was stored. As an example, in my case that was ~/.config/chromium/Default/databases/https_lesen.amazon.de_0/1
If you have installed the browser as snap package, you need to first check where that snap is mounted (e.g. with mount | grep snap).

Of course the ebook needs to first be downloaded for offline use: Go to the library view in the cloud reader App, right click the cover, click Download (if you don't have that option, you may not have the cloud reader extension installed).
If an ebook was opened in the browser, but not downloaded, the sqlite file will exist, but not contain the content necessary for export (and this script will thus fail).


This is much better then nothing, but far from perfect.

  • Generated files may become huge and cause troubles to some browsers. Storing images as standalone files may improve that.
  • Page numbers should probably be included (optionally?).
  • An option to extract only specific books would be useful for large libraries. Currently, it will extract all e-books found in the given file.
  • The script could support the user in finding the right sqlite file.