Using org-ql as powerfull filtering tool on top of org-roam to generate and publish any website

Hi,

I re-think about a potential great new application for org-ql,
by using it as a powerfull filter tool standing before any org publishing.

x * ( apply list of org-ql filters on huge .org file folders **----> generate --->** a new hierarchy org folder & files **----> with org-publish ---->** to static website), x = the number of web site you want to generate based on the same org files/folders.

As we discuss in org-roam, one of the difficulties of publishing org-roam galaxy of note (see the [issue](https://github.com/org-roam/org-roam/issues/911) and discourse discussion [here](https://org-roam.discourse.group/t/package-dedicated-to-workflows-publishing-digital-garden-on-web/536) ) was to :
- filtering note by tag
- filtering note by private or public status (possibly tagn but for gpg notes ?)
- decide in a rule if we want to keep or not notes when you have a "private-public backlink" between two notes ?
- how to re-route each tag/query to folder ?
- generate new content (index of files) based on query (see issue [here](https://github.com/alphapapa/org-ql/issues/180) ) ?

Using org-ql and org-mode tangle, all of this seems really possible, by
- creating new org content based on org-ql queries stored on folder/files
- routing content into new hierarchy of folder and files, by coupling org-ql query and weblorg routing ( https://github.com/emacs-love/weblorg perhaps, see issue here https://github.com/emacs-love/weblorg/issues/18 )?
- generate dynamic transclusion block (adding / syncing org-mode block into any other org-mode content dynamically, by using for example https://github.com/nobiot/org-transclusion)

With all of that i suppose that could be very easy to extract/build/publish illimited website on top of any org-mode folder, what do you think ? I'm really interested by this approach because that could help us to build very organic static website.

Exemples :
- generate a blog by creating a folder /blog that contain result of a org-ql query on any headers tagget as "blog"
- generate category index all page (as wikipedia do) by querying/grouping on category
- etc.