I'm not sure if there's a consensus in linguistic nomenclature about using the aforementioned prefix in naming the reconstructed languages.
As we all probably know, in linguistics, there's a custom of naming every unattested, hypothetical, reconstructed mother-language with by adding the prefix "proto-" to the family name.
What I am curious about is, is this a very strict rule and if one would find attestation of even one or two words from the reconstructed language, would it impose a necessity of dropping the prefix from the language name or calling it differently.
In addition, what about the names in the model of "Common Germanic", "Common Indo-Iranian"? Is there any distinction here between these and the "proto" type ? Frankly, I have encountered both of those and I am wondering if the choice of using one indicates anything.