If I read you correctly, your point is this: if you're prepared to assert P, you should be prepared to assert that you know that P. And the converse is even clearer: if you are willing to assert that you know that P, you're willing to assert that P is true. That's an interesting and important observation, but it doesn't show that the standard analysis of knowledge is circular. Suppose I'm prepared to assert that P. Do I actually know that P? That depends. Even if I'm prepared to make a sincere assertion -- and hence believe that P -- I might not really be justified or P might not actually be true. In either case, the classical analysis says, I don't actually know that P. The analysis of knowledge doesn't make any reference to what people are prepared to assert. On the contrary: it points out how there can be a gap between what we're prepared to assert and what we actually know.
We could turn this into a slogan: saying it's so is saying you know, but that doesn't mean you do.