Reference no: EM132310134

Leveraging static analysis tools to pinpoint security issues of software applications

The students need to select a well-known static analysis tool (e.g., FindBugs for Java, AndroBugs for Android) and apply it to analyse various software applications (e.g., open sourced projects from GitHub or Android APKs downloaded from markets). More details will be given during the lecture session.

The output of this assessment should include (1) a detailed report describing all the details involved and (2) an executable software package that could be leveraged to replicate the reported experiments.

Tasks

1. Choose a static analysis tool

There are quite a lot of static analyzers available in the community. Here I just list several well-known examples that the students can choose from. The student can still choose other tools including other language-based tools.

- Java: FindBugs
- Android: AndroBugs
- C++: RATS, Flawfinder

2. Understanding the selected tool

The students should explore the selected tool in various aspects, attempting to summarise the capabilities provided by the tool. One way to identify such capabilities could be to go through all the input options that the tool provides.

3. Finding target code to analysis
In this step, the students should choose at least one project (or one Android app) to perform their analysis and that project should contain security issues. Github is a great source that the students can leverage to finding target projects. The following screenshots demonstrate various projects that are mainly written in Java and C++, respectively.

4. Launch the selected analyzing tool to the selected repositories (or Android apps)
After selecting the to-be-tested projects, the next step is to launch the selected static tool to analyse the source code of those selected projects.

The students should clearly describe the setup of their experiments, e.g., how the selected tool is launched, what is the inputs provided? The students are also expected to justify the complexity of the selected projects. How big is the selected projects? Ideally, the more complex the projects selected (e.g., top 10 Java projects hosted on Github), the higher score the assessment will be.

5. Summarize the analyzing results of the a forementioned experiments
After launching the static tool on the selected projects, the tool should generate some results, ideally security issues related to the analysed projects. The students should then summarize those results in various means, attempting to represent those results in more understandable ways. Last but not the least, the students should also provide some insights that are learned from the experiments and could be useful for other code analysers.

Interesting directions that the students are encouraged to explore:
- Comparing the capability of different static analysers
- Comparing the security issues of different projects
- Comparing the different revisions (github releases, tags) of the same project and hence to understand the evolution of security issues.

Attachment:- Assessment Specifications.rar