Ah yeah two things:
(1) in the header of the action “Execute script”, you need to hit the “+” button to add a new variable (html_images). You should be able to find the html_images variable in the @-mentioning menu in the web page action
(2) The “deliverable_field_image_file_urls” variable needs to be changed to the field of your app. You can simply delete the variable name and start typing the name of your image field, the auto-completion should do the rest!
Thanks for documenting this useful scripting snippet. Here is the share link for anyone who wants to quickly duplicate an example app + automation of this into their organization: