Hacking Java Clients

Status

Released 14/1/2008

Hacking Java Clients

When performing a security assessment of client-server Java applications, it is sometimes necessary to modify the client component in order to properly understand and assess the security mechanisms in place. Typical examples are systems that employ a communication channel that can't be intercepted with tools such as the personal proxies (WebScarab, Paros, etc.). A convenient means of accessing the internals of a Java program is to have an interactive scripting environment (BeanShell, Jython, JRuby, Groovy, etc.) that exposes the internal objects and allows you to perform arbitrary operations on these objects. The following white paper outlines this technique.

There are a number of techniques that can be used to insert such an interpreter, these include:

1. Recompile the source code and include the interpreter into the app. (Of course, you'll need access to the source and a build environment)
2. Insert the interpreter using inheritance (as described in the white-paper mentioned above).
3. Insert the interpreter by directly manipulating the byte-code
4. Use this tool
5. Use a new feature implemented by Java 6