It's not an automatic clustered index. It's by default clustered, that's all. Trivial to have a nonclustered primary key or a clustered unique constraint

PK disallows nulls.

This is more a DB design concept than implementation. When you're doing a logical design, you identify the candidate keys (column or sets of columns that are unique). You then chose one as the primary key and the others become alternate keys. When implemented, the primary key is set as the primary key and the alternate keys should get unique constraints or unique indexes.