I just got ProcessGuard and I like it a lot. I've been running some leaktests to see how secure my system is, but there is a weird occurence. If I try it for the first time, I click "Yes" when ProcessGuard makes sure I want to run the program. However, there are sometimes no consequent requests from ZoneAlarm when the program tries to access the internet. Therefore one of the tests has been successful the first time. However when I ran it again, I wasnt prompted by PG (I had saved it as an allowed program) but then it failed because I was prompted by ZoneAlarm, and denied it access causing it to fail, and my security to pass. However the first time should be just as secure because in a real world situation, all it needs is one chance. Although I'm not just gonna go around allowing programs I've never heard of to execute and if it wasnt disguised I'd probably have blocked it from running in the first place, there are times this may not be the case. Here is the exact happening: I get TooLeaky, and run it. I allow and remember it in PG. I start the test. It is successful (and my security failed). I run the test again, and this time PG remembers it and doesnt prompt me. However I get a Dangerous Behavoir alert from ZoneAlarm informing me that it is trying to launch IE. Needless to say, I dont allow it, and I pass the test. However allowing a program to run, and then discovering dangerous behavoirs are different, and I want to know why I'm not prompted by PG and then alerted by ZoneAlarm once that actual program is launched, and if I can fix this.