Resolver for ambiguity in reference style in Links?

[pseudomonas]

I'm at a bit of a loss how to disambiguate the references in the Links resource. I'm starting with a reference and trying to get the text of the verse/sentence/pasuk/smallest-grained-text-string that the citation refers to.

Let's say there's a reference called Ploni on Tanakh, Book III, 5:2 then the filename of the schema file that I need to look up could be any of Ploni.json, Ploni_on_Tanakh.json or Ploni_on_Tanakh,_Book_III.json. I can do that recursively. The sentence referred in merged.json to could be any of
["on Tanakh"]["Book III"]["4"]["1"] (the 5,2 becomes 4,1 because usually the json is zero-based while the references are one-based, though sometimes there's a zeroeth verse called something like "Introduction")
["Tanakh"]["Book III"]["4"]["1"]
["Book III"]["4"]["1"]
["2"]["4"]["1"]
["4"]["1"]

Is there a parser/resolver for the links (short of installing the full software, Node, Mongo, database, web interface and all and using the web API)? I'm having problems with MongoDB.

[pseudomonas]

Addendum: I've been using the cltk-full file tree. Have I been doing it wrong?