You should make a program like xenu link sleuth that can actually handle millions of urls and not run out of memory.
I used to be a c# developer and have written something similar using the abot nuget package, but its way to clunky for my liking. However, it works for what I need it to.
Basically you can put in a list of domains (or possibly keywords and it will find domains) it should traverse the domains inner links and log all external links. The user should be able to set a maximum/minimum crawl depth and whether it should crawl the logged external links along with a min/max depth to crawl.
Multi threaded and keeping the UI responsive would be a must.
To top it off you could check to see if the logged external links (domains) are registered or not.
I would be willing to pay upwards of 90$ one time fee for something like that, but I think a more than fair price for most would be 30-40 range.