Certain elements such as VeDmBlockquoteNode are very unlikely to be appropriate targets for text matches. We should probably allow excluding/including node types.

We're already checking for elements that appear at the root of the document (instead of eg in captions or quotes) in T413421: [Suggestion] Detect if an internal link appears >1x in the same section.
This ticket was raised because it would also be practical in T413420: [Suggestion] Detect when a variety of English is being used that does not align with article's tag

• Discuss whether/when to do this
• Discuss whether to have include and/or exclude lists, and/or whether to have a simple "must be at root" boolean option
• Explore whether this could be a helper function any check can use

See also:

• A userscript which excludes using regexes