proxy http://proxy.cnn.com *.cnn.com/*will use the proxy http://proxy.cnn.com for all pages that match the pattern *.cnn.com/*. The default port for the proxy is 8080 as usual, but you can also define the port, like
proxy http://proxy.cnn.com:80 *.cnn.com/*
Webmirror will check the patterns in the order as they are defined in the RDF and if one matches the URL that proxy will be used for download. If none of the patterns match webmirror will try to download the page directly.