Virtually everyone uses Google or other search engines, but what most people don't know is that these search engines can perform advanced queries that are exploited to carry out successful cyberattacks.

For example, earlier this year, a cyberattack by suspected Iranian hackers made headlines when they used a simple technique called Google Dorking to access the computer system that controlled a water dam in New York. Google Dorking is readily available and has been used by hackers for many years to identify vulnerabilities and sensitive information accessible on the Internet.

Since its inception, the capabilities in Google Dorking have been added to other search engines, including Bing, Baidu, and Open Source Network Intelligence Tools (OSNIT) such as Shodan and Maltego.

Google Dorking, however, isn’t as simple as performing a traditional online search. It uses advanced operators in the Google search engine to locate specific information (e.g., version, file name) within search results. The basic syntax for using an advanced operator in Google is Operator_name: keyword

The use of advanced operators in Google is referred to as “Dorking” and the strings themselves are called “Google Dorks.” Dorks can be as basic as just one string, or they can be a more complex combination of multiple advanced operators in a single search string. Each Dork has a special meaning to the Google search engine that enables hackers and others to filter out unwanted results and significantly narrow down search results. For example, Google Dorks can be used to find administrator login pages, user names and passwords, vulnerabilities, sensitive documents, open ports, email lists, bank account details, and more.

Anyone with a computer and Internet access can easily learn about the availability of advanced operators on Wikipedia or via other public sources. Therefore, it’s not surprising that federal authorities say it is increasingly being used by hackers to identify computer vulnerabilities in the United States. The Department of Homeland Security and the FBI in 2014 issued a special security bulletin warning the commercial sector about the risks of Google Dorking.

The underlying threat associated with Google Dorking is that search engines are constantly crawling, indexing, and caching the Internet. While most of this indexed data is meant for public consumption, some is not and is unintentionally made “accessible” by search engines. As a result, a misconfigured intranet, or other confidential information resource, can easily lead to unintended information leakage.

Considering how easy it is for cybercriminals to access sensitive information via public search engines and security tools raises an important question: What can organizations do to minimize the risk of being hacked via Google Dorking?

The first step is to avoid putting sensitive information on the Internet. If unavoidable, assure that the data is password-protected and encrypted. In addition, make sure that websites and pages that contain sensitive information cannot be indexed by search engines. For example, GoogleUSPER provides tools to remove entire sites, individual URLs, cached copies, and directories from Google’s index. Another option is to use the robots.txt file to prevent search engines from indexing individual sites, and place it in the top-level directory of the Web server.

More important, organizations should implement routine Web vulnerability testing as part of standard security practices. In this context, Google Dorking can be a proactive security tool using online repositories like the Google Hacking Database (GHDB), which documents the expanding number of search terms for files containing user names, vulnerable servers, and even files containing passwords. The database provides access to Google Dorks contained in thousands of exploit entries. The direct mapping between Google Dorks and publicly available data allows security professionals to more rapidly determine if a particular web application contains these exploits.

The Google Dorking phenomenon once again underscores how organizations must not only test for vulnerabilities, but also assess whether they can be exploited, and what risks they represent. This is best achieved when vulnerability assessment, penetration test, and a cyber-risk analysis are performed hand in hand.

Related Content: