A general note on phenotype
I think that the meaning of phenotype has become more complicated over time compared to what it was when it was first coined. As you said, in simple words, a phenotype is an observable characteristic of an organism. However, for Mendel, an observable characteristic would have been only an obviously gross change in the morphology whereas with the current state of technology we can observe the assembly of macromolecules inside a cell. We can also study the physiology of an organism/tissue/cell in real time using various sensors. We can also observe the gene expression profile using different technologies (with different levels of sensitivity and resolution). The question now is, can you call such a characteristic (for e.g. an assembly of some complex in a cell observed by cryo-EM) a phenotype? In other words "what is an observable"?
Coming back to your question:
Can it be said that proteins determine phenotypic traits?
Are they the sole determinants of phenotypic traits?
No. A phenotype is usually determined by many factors. Many genes are involved, which may code for proteins as well as functional non-coding RNAs. However, if you make a comparison, proteins seem to be more versatile than ncRNAs (also most of them are involved in direct action). So you may say that proteins are the primary determinants of most phenotypic traits, but I would generally advise that such generalizations should be used cautiously. I'll give a counter example in favour of RNAs. If there is a mutation in a tRNA or a rRNA, you'll see a great effect on the phenotype.