Automated Web Scanners Bring The Noise

One fish, two fish, red fish, skipfish...huh? That was my initial thought. Skipfish is definitely an interesting name for a Web application security scanner. It sounds like it came straight out of a Dr. Seuss book, but instead it's an awesome new tool from Michal Zalewski and Google.Web security tools aren't anything new to Michal, who previously wrote ratproxy, a passive semiautomated Web application security-auditing tool that acts as a proxy to your Web browser while performing checks. Initial reports state that skipfish is fast but noisy. After some testing of my own, it is screaming fast. And as you might expect from an automated scanner, it is also very noisy.

So here's my question to you: Isn't that what you'd expect from an automated Web application scanner? I've used numerous open-source Web scanning tools from simple scanners, like Nikto, to more powerful and complex tools, like w3af. I even use a commercial solution. Whenever you're scanning and using brute-force file and directory discovery, it's going to be noisy, with tens to hundreds of thousands of 404 (file missing) error messages as it's attempting to discovery interesting locations not linked from the main site.

After reading through some of the comments in blogs and on Twitter about Skipfish's "noisiness," I started comparing a couple of other tools that have brute-force file and directory discovery features. Nikto scans for more than 6100 dangerous files/CGIs and other checks. The documentation even states that it is not designed to be overly stealthy. No big surprise there.

In looking at the "dictionaries" included with skipfish, the complete config includes 1,802 "words" that will get combined with 92 "extensions" to become one monster list of files and directories that get probed.

DirBuster is another well-known discovery brute-forcing tool. I've only used it once in Web penetration testing training session so I forgot just how cool it was because of the way the directory lists are created. As the tool's overview page says, "tools of this nature are often as only good as the directory and file list they come with. A different approach was taken to generating this. The list was generated from scratch, by crawling the Internet and collecting the directory and files that are actually used by developers!"

That just rocks, and it gives DirBuster the largest list I've come across in a Web scanner. My next test will be to integrate the two and see if I can make Skipfish's brute-forcing capability even better, although at some point I can foresee it getting possibly too long making the scan unbearingly long.

Skipfish is definitely a cool tool with a bright future, but don't forget what it is and what it does. Its inherent nature is going to cause it to be a noisy tool because it scans a Web app looking for certain files and all those requests will (or should) be logged, which can generate some hefty log files.

Quick update: The documentation for the dictionaries included with Skipfish discusses the DirBuster approach at the very bottom of the document and why Skipfish doesn't use it. Basically, "keywords related to popular websites and brands are heavily overrepresented," and "some of the most interesting security-related keywords are not commonly indexed." So, I likely won't bother testing DirBuster's list with Skipfish and will trust Michals's design decision. :)

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.