I was thinking of something like this. You have a test tool that then spawns another EXE which actually triggers AutoSandbox. As far as i understand the sandbox and its chain of sandboxing, in this case the main tool would not be sandboxed but the newly spawned EXE would be. Then the main test tool would check for the presence of that file and registry value. If found, Sandbox is not working. If not found, sandbox is working fine.
It's just that you'd have to figure out how to make timings and stuff like that to properly connect spawning and checking part so they would be properly correct and not try to check before it would actually spawn the new file/reg value. In theory you should only check the physical locations and if test data is not found there, it's fine.
Well yes, that would be possible (as for the timings, it would be best simply to wait for the spawned process to terminate.
However, keep in mind that it's more a test of the sandbox than autosandbox - in real life, you don't have control of whether you get the autosandbox offer for the particular file or not.
(Basides, I'm kinda scared when I read about "writing tutorials about using this tool"... it's just an artificial example that triggers the autosandboxing heuristics, no big deal - while it seems to be handled like some complex pen-testing tool
)
Can somebody please tell me how you open the file in autosandbox.
Thank you
There is no "how" - you just execute it.