How is static object state shared / set between MSpec contexts?

While writing MSpec BDD tests, I ran into a scenario where the test I expected to get passed, but only when I ran all my tests. When I ran the test individually, it failed as expected. After some research, I found that some state that was set in the previous test was not reset until the second test was done and this caused the second test to pass when I expected it to fail ... The following contrived code reproduces the scenario:

public class ContextBase
{
    protected static object state;
}

public class Context_a : ContextBase
{
    Establish context = () => { state = new object(); };

    It should_set_state = () => state.ShouldNotBeNull();
}

public class Context_b : ContextBase
{
    Establish context = () => {  };

    It should_set_state = () => state.ShouldNotBeNull();
}

      

        
        
        
      

    

Both of these tests pass because Context_a runs before Context_b, and at the time Context_B is running, the state that was set to Context_A is still set. If you run Context_B in isolation then the test fails because the state is not set. Interestingly, if you remove the empty Set statement from Context_B, Context_B will always fail as expected.

I am relatively new to MSpec and this behavior surprised me. I assumed such a state would be reset between the execution of each context. I may have missed something ... Am I building these tests correctly? If MSpec doesn't support reset in this way automatically between contexts, what strategy should I use to ensure that the state is reset in cases like my example? Should I put a New label in the ContextBase class that sets all state fields to zero?