Unbeknown to me, the other day I started receiving a bunch of these emails from twitter reporting that some of the URLs I have posted in my savory tweets pointed to malware domains.

My first reaction was panic; I turned off the twitter posting functionality.

But then I got another thought in my head.

The emails from twitter don't suggest whether or not they are marking my account for possible blacklisting; they're only removing the tweet.

savory has the ability to read emails...see where I'm going with this?

So I decided to give savory the ability to use twitter's own service to enhance my service : ) It works like this.

  1. I tweet the URL in my post to twitter.</li>
  2. I save the bit.ly URL in a "bitly" key in savory</li>
  3. I create a procmail filter for mail from twitter that resembles their message</li>
  4. If and when I receive an email from twitter, I call the savory API to retrieve the doc that has the bitly key with the value sent in the twitter email</li>
  5. I add another key to the document, "twitter_detect", with a value of "yes" that signifies whether this URL was detected by twitter</li>
    </ol>
    Beware, evil genius in the making.