I would that the script takes the content of all posts of a site, translates three times the content and publish it in a wordpress cms.
The script must take only the text and links and also a picture of the post without advertising.
All sites are in html. All sites have a sitemap.
This script must be run every day and not have to publish the same post, so all the posts that have the same name will be discarded, all the posts that have a name that contains similar words will be put in a state of approval.
The grab will be done following the sitemap and not rss.