I've dealt with a lot of different types of bots. The communication channels among them have varied from unsophisticated IRC command and control (C&C) servers to advanced peer-to-peer (P2P) protocols. For botnet herders, the challenge is flying under the radar of network security professionals who are monitoring their networks and looking for anomalies. The infosec pros who know their networks inside and out are likely to pick up on strange protocols pretty quickly -- which is one of the reasons HTTP bots have been so effective.Blocking HTTP is impractical for many organizations, opening up the opportunity for bots to reach out to their HTTP C&C servers. Zeus and Conficker are two examples of bots that have used HTTP. Malware researchers have published a list of known Zeus HTTP C&Cs, and that's where Conficker has upped the ante, making it much harder to track because it can check a huge list of domains generated daily and still communicate via P2P.

What about bots that use social networking sites? There have been a few discussions and examples of using Blogger in the past and, more recently, Twitter, but very few proof of concepts -- until this morning. Robin Wood, from digininja.org, posted an e-mail to the PaulDotCom mailing list about a Twitter-based bot, called TwitterBot, that he wrote in Ruby.

Robin's example is simple, but gives a glimpse of what could be done. In his example, you create an account that the bot follows. When you want the bot to do something, you post a "tweet" to the C&C Twitter account. The bot will then execute that command upon its next check-in. Very cool stuff. For more info, check out Robin's page.

Defense against bots using social networks is easy if you can simply block all social networks. But that might not be an option for companies that are increasingly using social networking to spread their marketing message. For example, it's not uncommon for the CEO of a tech company to blog, or for the marketing team to use Twitter to discuss new products.

Trying to defend against HTTP bots gets even trickier when you realize that social networking sites aren't the only public avenue to post commands. Consider sites, like Amazon, that let you post product reviews. It would very easy to post a seemingly innocuous comment about a product that turned out to have an embedded command in it.

Is it time to rethink letting your employees have Internet access? I know that seems drastic, but how many of your employees really and truly need Web access to do their jobs?

John H. Sawyer is a senior security engineer on the IT Security Team at the University of Florida. The views and opinions expressed in this blog are his own and do not represent the views and opinions of the UF IT Security Team or the University of Florida. When John's not fighting flaming, malware-infested machines or performing autopsies on blitzed boxes, he can usually be found hanging with his family, bouncing a baby on one knee and balancing a laptop on the other. Special to Dark Reading.