The TODO part is where I have to put my code. But everything I read for XML::RSS::LibXML doesn't let me pull out "item" tag.

#!/usr/bin/perl
use File::Path; use Data::Dumper;
use LWP::UserAgent;
use XML::RSS::LibXML;
use POSIX qw(strftime);
use Time::HiRes qw(gettimeofday tv_interval);
my $client = LWP::UserAgent->new();
my ($fh, $feed, $feed_title, $count, $node);
my $rss = XML::RSS::LibXML->new;
my $website_name = "usnews";
my $url = "http://www.usnews.com/rss/health-news/index.rss";

$firstListing = 1;
while (1)
{
    if ( $website_name eq "" )
    { next; };

    print "polling: $website_name url: $url\n";
    $capture = $client->get("$url", ":content_file" => "/tmp/.rss_down
+load_file") || die"$!\n";
    $rss->parsefile('/tmp/.rss_download_file');

    print "channel: $rss->{channel}->{title}\n";
    @curListOfItems = ();
    foreach my $item (@{ $rss->{items} })
    {
        my $node_link = $item->{link};

        if (defined $node_link)
        {
            $curItem=$node_link ."\n";
            push (@curListOfItems, $curItem);
        }
    }
    if ($#prevListOfItems != -1 )
    {
        # @newlyAddedLinks will be latest in curListOfItems and not in
+ @prevListOfItems
        @newlyAddedLinks=grep!${{map{$_,1}@prevListOfItems}}{$_},@curL
+istOfItems;

        foreach my $l (@newlyAddedLinks)
        {
            my $fileName=getFileName();
            $fileName="/tmp/.$website_name\_${fileName}";
            my $capture = $client->get("$l", ":content_file" => "$file
+Name");
            # TODO: Pull out the current Item tag ( <item> .....</item
+> )
        }
        print "Getting1 $filename\n";
    }
    elsif ( $firstListing == 1)
    {
        print "Getting2 $filename\n";
        foreach my $l (@curListOfItems)
        {
            my $fileName=getFileName();
            $fileName="/tmp/.$website_name\_${fileName}";
            my $capture = $client->get("$l", ":content_file" => "$file
+Name");
            # TODO: Pull out the current Item tag ( <item> .....</item
+> )
        }
        $firstListing = 0;
    }
    @prevListOfItems = @curListOfItems;
    open OUT_FILE, "> /tmp/.$website_name" || die "could not open file
+ $!";
    print OUT_FILE "@prevListOfItems";
    close OUT_FILE;

    sleep 1;
}
sub getFileName
{
    my ($seconds, $microseconds) = gettimeofday();
    my $padded_usecs = sprintf ('%06d', $microseconds);
    my ($logType, $str1, $str2) = split ('\|',$LogElement);
    $todaysDate = strftime "%d", localtime;
    $currentDateTime = strftime "%Y:%m:%d:%H:%M:%S", localtime;
    ($Year,$Month,$Date,$Hour,$Minute,$Seconds) = split /:/, $currentD
+ateTime;
    $curYear = sprintf ('%04d', $Year);
    $curMonth = sprintf ('%02d', $Month);
    $curHour = sprintf ('%02d', $Hour);
    $curMinute = sprintf ('%02d', $Minute);
    $curDate = sprintf ('%02d', $Date);
    $curSec = sprintf ('%02d', $Seconds);
    my $fname = "${curYear}${curMonth}${curMonth}${curHour}${curMinute
+}${curSec}.html";
    return "$fname";

}
[download]

So, you're parsing an RSS file and then, for each item, you are fetching the link. What you get back is HTML not RSS so no, I don't think you'll get far trying to process the links with XML::RSS::LibXML.

I'm not sure what you plan to do with the HTML documents but you already have XML::LibXML loaded into RAM so you could use it to parse the HTML:

use XML::LibXML;

my $dom = XML::LibXML->load_html(
    location => $fileName,
    recover => 1,  # handle marginal HTML
);
print $dom->toString;
[download]

The parser options for load_html are documented in XML::LibXML::Parser.

I just keep polling for the same rss feed and compare the Link from last poll and when ever new Links or Item has been added I only want to parse the Item tag of the New Item.