That's pretty much it. There needs to be a 'first wall' or structural vessel that encases the plasma, which operates in a vacuum. Water (or similar fluid) would simply evaporate in the vacuum and quench the plasma. And the first wall materials will eventually become activated.
Stainless steels are affected by (n,p), (n,α) and the p's and α's accumulate interstially, although H forms metal hydrides. The other effect is clusters of dislocations, which can be annealed out. Ni-58 can undergo an (n,p) reaction which produces Co-58, or Ni-58 absorbs the n, and produces Ni-59 which decays to Co-59, which can absorb a neutron and become Co-60.
There has been a big research project at ORNL on first wall materials, which also has relevancy to traditional steel pressure vessels and core structures of LWRs and fast reactors.
Lithium could actually flow inside the first wall, but that's tricky. The issue is the n(6Li,α)T and the effect of the α on the plasma, which is why it would be necessary to confine it outside the first wall. There's also the issue of confining T, and recovering it. Li-7 can be used as a coolant since it has a lower cross-section for n-absorption.
An aneutronic reaction e.g. d+3He would be ideal, but it's problematic given the cost and low abundance of 3He.