The docs are right, CB is not the right tool for this.
If you really want to list every word in the text (i.e. every surface form) then this simply an exercise in creating the word list for the project and then extracting the context for the first five references. I’m not aware of a tool that will do that but it’s not that hard in principle provided you can write the code you need to do it.
If by ‘every word’ you mean every (disambiguated?) lemma then it gets a little harder. Again, there isn’t a tool designed to do this directly but if your project has worked with the interlinear to parse the wordlist into stems and morphs then you may have at least some of the raw material to hand.
Message me off list if you want talk in more detail. Blessings, anon907613 R.