I don't use a Web browser to download software. I use "GNU wget" from the Free Software Foundation. It's a "command line" utility with lots of options. Wget preserves the time stamp on the downloaded file, so you can tell when the file was created, not when you finished downloading it. It's got a -c option to continue an interrupted download where you left off, and it works even if you continue from a different mirror source. It's got a --limit-rate= option so you can tell it not to choke your bandwidth while you do other stuff. It's been ported to MS-Windows and there's an add-on graphical user interface.
Most software worth downloading is distributed with an MD5 checksum. You download the software file, calculate the MD5 checksum at your end, and compare it against the MD5 sum at the source. This will protect you from bad downloads. If you check the sum at the original source it will protect you from a corrupted copy on a mirror site. It's really hard to make an altered version of a file with the same MD5 sum *and* exactly the same length, so you can be pretty sure you're not getting a fake. On GNU/Linux you have an "md5sum" program for calculating the checksum at your end. On MS-Windows there are lots of programs to do it.