Not saying that Perl couldn't do it too... but maybe a simple recursive grep is sufficient. Something like

grep -r searchword /path/to/htdocs/
[download]

(Of course that would search in the HTML markup etc. as well, not just the rendered page content... — Update: if that's a problem, you'd have to parse the HTML, for example using modules such as HTML::TokeParser, HTML::TokeParser::Simple or HTML::Parser. To recurse through the document tree, you could use File::Find)

Update 2: Maybe you could save yourself a lot of work by using some existing software like ht://Dig to index your entire pages, and then issue a simple search if you want to know which documents contain a particular term...