I think I understand what you’re working towards Chris and it’s very interesting. One of the problems I have with the current Post Kinds plugin is it only works with Microformats. When a site is not equipped with MF2 (and a lot of them aren’t) the plugin can’t parse any useful information from the site in question. When I want to reply, bookmark, favorite or anything, I need to be very aware I check the metadata of the site. Which is kind of weird, since there are lots of fallbacks when there is no MF2 available. It could trickle down as follows: MF2 > Open Graph > Meta tags of title and description. If this could be fixed within Post Kinds or the solution you’re working on, that would be great!
Footer
Wat is dit?
Digging the Digital is de digital garden of commonplace book van Frank Meeuwsen. Onderwerpen variëren van indieweb tot nieuwsbrieven, bloggen, muziek en opvallende gebeurtenissen op het internet. 
Frank, it may seem that Post Kinds only works well with microformats, but this isn’t the case. It’s running some more sophisticated code based on Parse This which looks for microformats/HTML first and then it drops back to schema.org, JSON-LD, and then OGP metadata as a back up. When Post Kinds doesn’t find or return much data, it’s because the site is providing none of these.
If you’re finding posts that do have some sort of mark up, but aren’t parsing properly, ping David Shanske and let him know.