I would like to think the definition of life involves a Russian doll of attributes. I don't have the technical background to list everything, but it would seem that replication is the most essential attribute, followed by variation, followed by metabolism.
Drawing a fixed line in the sand and saying this list of attributes is required to call something living seems to me more of a political or religious statement than a scientific one.
I would also argue that on earth at least, life is a community that supports a lot of parisitic entities. I am not convinced that viruses should be excluded from the definition of life simply because they require a host, any more than a fetus or profoundly retarded person should be classified a non-human because it can't request an attorney.
It's more complicated than it seems at first. Is a sperm cell living? An unfertilized egg? A fern spore? (I would have added lichens, but I think in all cases studied, the fungi and algae involved can live separately.)
Mostly the boundary isn't important outside politics or religion as you point out. A koala cannot live outside it's few trees-of-interest nor can a panda live outside restricted areas. Tubeworms have to live near black smokers.
You could call a boiling pot of water living if you want to. That does not make it so. A virus needs a host not just to survive, it needs a host before it can do anything. Any parasite that is alive, replicates by itself and metabolizes by itself. It may require food and some components from its host to complete its life cycle, but it nevertheless can do something without a host. And your strawman human, is non-existent. A human is not defined by the ability to request a lawyer nor to build strawmen.