The only area for improvement I would say is that the false positives are nearly zero; everything is mostly like 99 to 99.99% or we can say 100% accurate. There were a few areas for improvement just from the last time I saw; I think the user experience had a little problem. We wanted to have certain reports based on our kind of scenario, but the tool did not allow us to create custom reports. We had asked for some facility and some ability for us to create some custom reports. That would be awesome if they allow us to create custom reports the way we wanted. There is one small area which I don't know whether we should call a tool limitation or a wish list; if I use a library and I don't use all the capabilities of the library but only a portion of it and that portion is not vulnerable, but there is a component which is outdated, that is a problem, even though I don't use that component. Mend.io will discover there is a problem in the whole library; that is correct. That's a valid discovery, but in my case, for example, if I don't use that particular portion, then it actually is not making sense for me, but that's not a limitation of Mend.io; I think that's a general problem with any tool in the market because no tool in the market will actually know what portion of the code I'm actually using from that particular library if it is vulnerable or not.
