I'm not sure I quite understand exactly what you're looking for, so correct me if I'm wrong... What I think you want is the Cwd module that comes included with the standard distribution. If you have /dir1/cgi which calls cwd(), and a symlink to it in /dir2/cgi, executing /dir2/cgi will have cwd() return /dir2, so if you use that in the main cgi, you can always determine the directory that it is linked from. For the url part of the question, I don't think you can get what you want through HTTP_REFERER, as that will contain the url that the person came from, not the url of the cgi that is linked to. The best way to go about it would probably be to have some sort of base url inside the perl script, to which part of the directory can be appended. This also prevents any potential security problems of trusting that the user isn't messing with their Referer header.