You have all heard of j0hnny l0ng by now. The idea is that there are a lot of google queries that you can do to attack a target. The truth is the work he has done is very impressive and it is hours of fun to go through his database and see how dumb everyone is. What some of us that secure places for a living are wondering is, is *my* site vulnerable to any of these? Enter the Google Vulnerability Scanning Engine (Gessus?). This is my idea: to take the list of queries as given by j0hnny’s database, modify them slightly to restrict the search to only your domain or IP space as relevant, and return the results in a form that is similar to Nessus output. The great thing about j0hnny’s site is that the community has ranked each search in terms of effectiveness, which can translate almost directly into severity.
So I visualize the output of the tool to be very much like a nessus run output, an html page ranked by severity perhaps by ip or CIDR block. You click and drill down into one of the ‘alerts’ and you can look at the actual google search results in thier entirety. j0hnny has also taken the liberty of including a description field under each query, and that information would be used to add detail to the output, so that the examiner could quickly determine what the actual search results mean.
Ok the two problems: First, j0hnny, as far as I can tell, doesn’t provide a nice feed of all of his queries. We are faced with having to do an ugly and not very nice site rip to gather the data we need. It would be great if j0hnny would provide his database in an easy to use form (xml?).
Secondly, if we want to play nice with google, we’ll use the google API to do our searching. This is all well and good until we realize that we are limited to 1000 searches a day with this mechanism. To do this legally, we’ll have to batch the queries into several days. This shouldn’t be too much of a problem, though, cuz this data would only need to be gathered at most once per week.
So what do you think?
Posted in Uncategorized