It could be reasonable to collect some past examples of such proxies. This would help in assessing whether something like this is technically feasible. Also I assume that if something like this is implemented at some point, we'll need to prepare list of regular expressions for such automatic rewrites - so it is good which proxies we are concerned with. (The post is CW, feel free to add more examples.)

 * `proxy2.lib.umanitoba.ca` (http://journals.cambridge.org.proxy2.lib.umanitoba.ca/action/displayAbstract?fromPage=online&aid=7629848 vs. http://journals.cambridge.org/action/displayAbstract?fromPage=online&aid=7629848 in [this answer](https://mathoverflow.net/revisions/161852/1)
 * proxy.library.cornell.edu (http://www.numdam.org.proxy.library.cornell.edu/item?id=CM_1983__49_3_405_0 vs. http://www.numdam.org/item?id=CM_1983__49_3_405_0 in [this question](https://mathoverflow.net/revisions/183160/4))
 * `ezp-prod1.hul.harvard.edu` (http://dx.doi.org.ezp-prod1.hul.harvard.edu/10.1119/1.11104 vs. http://dx.doi.org/10.1119/1.11104 in [this answer](https://mathoverflow.net/revisions/208477/1))
 * `libproxy.mit.edu` (http://www.sciencedirect.com.libproxy.mit.edu/science/article/pii/0012365X9500120L vs. http://www.sciencedirect.com/science/article/pii/0012365X9500120L in [this question](https://mathoverflow.net/revisions/179138/2))
 * `libproxy.smith.edu` (http://www.jstor.org.libproxy.smith.edu:2048/stable/2031721 vs. http://www.jstor.org/stable/2031721 in [this question](https://mathoverflow.net/revisions/218884/3))
 * `ezproxy.library.nyu.edu` (http://ezproxy.library.nyu.edu:2233/mathscinet-getitem?mr=1042048 vs. http://www.ams.org/mathscinet-getitem?mr=1042048 in [this answer](https://mathoverflow.net/revisions/167652/2))
 * `lib-ezproxy.tamu.edu` (http://dx.doi.org.lib-ezproxy.tamu.edu:2048/10.4153/CJM-1972-118-x vs. http://dx.doi.org/10.4153/CJM-1972-118-x in [this answer](https://mathoverflow.net/revisions/79221/1))
 * `ezproxy.tcu.edu` (http://www.ams.org.ezproxy.tcu.edu/mathscinet-getitem?mr=1650341 vs. http://www.ams.org/mathscinet-getitem?mr=1650341 in [this answer](https://mathoverflow.net/revisions/27319/2); http://dx.doi.org.ezproxy.tcu.edu/10.1515/crll.1907.132.270 vs. https://doi.org/10.1515/crll.1907.132.270 in [this answer](https://mathoverflow.net/revisions/222364/3)) 
 * `proxy.uchicago.edu` (https://search-proquest-com.proxy.uchicago.edu/docview/1424274072 vs. https://search.proquest.com/docview/1424274072 in [this answer](https://mathoverflow.net/revisions/289453/1))
 * `proxy.lib.umich.edu` (http://www.jstor.org.proxy.lib.umich.edu/stable/2372821 vs. http://www.jstor.org/stable/2372821 in [this answer](https://mathoverflow.net/revisions/234841/2); http://jlms.oxfordjournals.org.proxy.lib.umich.edu/content/s2-15/1/134 vs. http://jlms.oxfordjournals.org/content/s2-15/1/134 in [this answer](https://mathoverflow.net/revisions/214777/3)) 
 * `ezproxy.is.ed.ac.uk` (http://www.ams.org.ezproxy.is.ed.ac.uk/mathscinet-getitem?mr=867684 vs. http://www.ams.org/mathscinet-getitem?mr=867684 in [this answer](https://mathoverflow.net/revisions/215941/1))