The inclusion of | "n/a" is for some error handling in cases where the computer name doesn’t exist so that it will return ‘n/a’ for that field instead of erroring out. See Pipe Character in Relevance for some more examples.
As to a filter, it sounds like you want to filter on the value of the results, correct?
That might look something like:
(name of computer of it | "n/a", values whose (it as lowercase contains "spooler") of it) of results of bes properties whose (name of it = "Services" AND name of source analysis of it = "TR - Services")