When users cite content from The Wikipedia Library which they access via our proxy server, they might simply copy the URL in their address bar, which will include our EZProxy URL. We could invest in a bot to clean up these URLs, replacing them with the accessible versions.

For Rock's Backpages, as an example, this looks like https://www-rocksbackpages-com.wikipedialibrary.idm.oclc.org/Library/Article/camel-over-the-moon. For any user who isn't authorized to access Rock's Backpages via TWL, this URL will result in a Library Card authentication error. Instead, the user should have added the direct Rock's Backpages URL; https://www.rocksbackpages.com/Library/Article/camel-over-the-moon, which displays an abstract and the option to login with various other methods.

We could explore a number of options for solving this problem:

• Providing users with more guidance about how to avoid adding a proxied URL.
• Creating a bot from scratch capable of replacing a proxied URL with its original form.
  • This could also potentially un-proxy URLs from other libraries.
• Work with an existing bot (such as User:Citation bot) to integrate this workflow.

All OCLC EZPRoxy proxied URLs from the library on the English Wikipedia can be viewed at https://en.wikipedia.org/w/index.php?target=https%3A%2F%2F*.wikipedialibrary.idm.oclc.org&title=Special%3ALinkSearch

This problem isn't just limited to The Wikipedia Library - https://en.wikipedia.org/w/index.php?title=Special:LinkSearch&limit=500&offset=0&target=https%3A%2F%2F%2A.idm.oclc.org shows that there are hundreds of URLs added via access from other libraries using OCLC EZProxy.