In addition to deceptive actions such as redirecting to other sites, malicious content can install crimeware on a user’s computer through a web browser vulnerability or by social engineering, such as asking a user to download and install anti-virus software that actually contains crimeware.
There are three primary classes of content injection attacks, each of which has many possible variations:
- Hackers can compromise a server through a security vulnerability and replace or augment the legitimate content with malicious content.
- Crimeware can be inserted into a site through a cross-site scripting vulnerability.
- Malicious actions can be performed on a site through an SQL injection vulnerability.
Cross-site scripting and SQL injection are propagated through two different primary vectors. In one vector, malicious content is injected into data stored on a legitimate web server, which a victim is exposed to. In the other vector, malicious content is embedded into a URL that the user visits when he or she clicks on a link. This is commonly a URL that includes components that will be displayed on screen or used as part of a database query, such as an argument to a search function.