Sandboxie is a very popular and useful program for handling suspicious executables on Windows. Due to flawed security properties of Windows systems (eg. compared to Linuxes), it is hard to rely on the OS alone for containing threats, yet a full blown VM can have prohibitive performance cost. Sandboxie seems to offer a convenient middle ground, largely isolating processes without incurring the heavy cost of hardware emu/simu-lation.

However, how secure is Sandboxie, really? I am aware that even full VMs are not perfect, and exploits exist that allow a process inside to escalate into the host OS. Obviously these exploits are much rarer and more complicated to actually do, hence why people actually use VMs for security. Does Sandboxie offer comparable security to a VM, or even something like Linux jails? Or is it fraught with security holes due to being deployed on inherently inescure system (Windows) and being closed-source? What are the most egregious, known vulnerabilities in Sandboxie that have been observed in the wild?

  • Here is an interesting debate from GRC about that topic: media.GRC.com/sn/SN-172.mp3
    – Mirsad
    Mar 31, 2016 at 5:06
  • @Mirsad GRC is a snakeoil salesman. I would not listen to him.
    – forest
    Oct 18, 2018 at 7:14

1 Answer 1


It's decent! After all nothing is secure...

By that I mean:

In 2013 in a public audit it was broken ONLY with targeted malware that affected the OS Kernel, OS User Mode at an Sandbox Bypass level and Key logging, Remote Webcam/MIC Access, Clipboard Hijack, Screen Scraping, Steal Files (in some cases), Network Shares Access at an Sandbox Leakage level. Those problems were detected in 2013 and since then there were many security and optimization improvements.

You can read there last public security report at: https://bromiumlabs.files.wordpress.com/2013/07/application_sandboxes_a_pen_tester_s_perspective2.pdf


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .