As you already know, inheriting a legacy codebase may be a daunting task.

We stated that legacy code is code without tests, so the first step in dealing with it is to create tests to help you preserve the same functionality during the process. Unfortunately, creating tests is not always as easy as it sounds. Many times, legacy code is tightly coupled and presents other symptoms that show a poor design or at least a lack of interest in the code's quality in the past. Worry not: you can perform some of the tedious tasks step by step, as shown in http://martinfowler.com/bliki/ParallelChange.html. Moreover, it is also well known that software development is a learning process. Working code is a side effect. Therefore, the most important part is to learn more about the codebase, to be able to modify it with security. Please visit http://www.slideshare.net/ziobrando/model-storming for more information.

Finally, we encourage you to read Michael Feathers book called Working Effectively with Legacy Code. It has plenty of techniques for this kind of codebase, and as a result is very useful for understanding the whole process.