How can we defend ourselves against the new social sites bot hackers?

In the March 4, USA Today newspaper is feature story on computer is sending you a Facebook or Twitter “friend” message that has an embedded bot program that then records all your key strokes. upon hacking your list of “friends” they then start the infection of ever more computers. They are especially targeting social network access from company computers and once in with internal company password access data, finance, patents etc. How do we private home users diminish this bot threat? How to do so, when the message we get is in fact from a “friend” who’s screen name, avatar, password have previously been stolen and appear to be legitimate.