I agree. The current implementation does not make sense.

If I create a page in a subfolder, for example in /home/example, the associated page will be /home/example/home.md.

However, if I click the breadcrumbs and go up one level to /home/example, then Wikijs tries to load the page /home/example.md, which does not exist.

Perhaps it would make more sense to either have the new subpage create both the /home/example/ folder and the /home/example.md page instead of creating /home/example/home.md ?

Using Matěj example below, clicking on https://docs.requarks.io/en/guide gives you a page that does not exist.

You can create a page https://docs.requarks.io/en/guide and then the breadcrumbs link will work, but what to put in the page? It would be nice if you could have a simple page with a list of the pages in the subfolder(s) included in the body of the page.

In something like Wordpress, you can include a list of Categories within the body of a page for example. It would be nice if there was a shortcode which would display subfolders/pages and then having a simple page for each folder of your hierarchy would make the breadcrumb links work.

I agree that this is a problem - what is a benefit of an auto-created folder that doesn't contain anything?

I don't like the proposed solution though - if a link would not be clickable, then you end up with different breadcrumbs with sometimes clickable path components and sometimes not clickable ones, which is confusing.

What about having an "autoindex" functionality that would generate list of child entries when such non-existent page is accessed?

To illustrate the original issue, the description of the folder structure a https://docs.requarks.io/en/guide/structure demonstrates the Home/guide/structure breadcrumb. Clicking on "guide" takes you to https://docs.requarks.io/en/guide that doesn't exist. Boo!