We consider the question of how a conspiring subgroup of peers in a p2p network can find each other and communicate without provoking suspicion among regular peers or an authority that monitors the network. In particular, we look at the problem of how a conspirer can broadcast a message secretly to all fellow conspirers. As a subproblem of independent interest, we study the problem of how a conspirer can safely determine a connected peer's type, i.e., learning whether the connected peer is a conspirer or a regular peer without giving away its own type in the latter case. For several levels of monitoring, we propose distributed and efficient algorithms that transmit hidden information by varying the block request sequence meaningfully. We find that a p2p protocol offers several steganographic channels through which hidden information can be transmitted, and p2p networks are susceptible to hidden communication even if they are completely monitored. © 2011 IEEE.