Either find the links, get them, save them, or inject something like this and call it from the page for each target you've identified, or submit a patch to add the required functionality to this module, or choose something else to achieve your goal. Unless you need JavaScript there should be alternatives, but your post lacks enough detail to expand on that.

The links for these files

<a class="txt" href="file.txt"> Text File </a>
[download]

can be obtained using the WMC instance by doing something like this

my @links = $mech->find_all_links( text_contains => 'some description 
+etc... ' );
my @urls = map { $_->[0] } @links;
[download]

In the case of WWW::Mechanize (WM) you can simply download the files using this code

for my $foo (@urls)
{
   my $filename = '/path/'.$foo;
   $mech->get($foo, ':content_file'=>$filename);
}
[download]

Unfortunately, this function does not work with WWW::Mechanize::Chrome (WMC). I hope the Author of WMC can shed some light on this or provide a patch. Thank you.