Background: In another of my questions, Ether brought up the idea to check links in an answer against the WOT database. Web of Trust is a community driven project, with the goal to make browsing safer by utilizing a simple reputation and comment system. It has an API-Interface which can be used for requests.
Goals: Check links for a bad or low reputation when writing or sending an answer/question. This would increase the safety for unexperienced users and also the reliability of external sources.
Upsides: The clear benefit is that no harmful link could slip-in (intended or unintended), and unexperienced users are...ahm...safer. Another upside is that most spammers would be caught, because those sites already have a low reputation, though, it wouldn't stop them from posting I guess, but at least slow them down or make them stick out immediately.
Downsides: You'd need an off-site check against a third-party database, which needs some bandwidth and time...and of course time to implement and some maintenance. May sound like nothing big, but I understand that this can be a real deal-breaker.
Possible Usages: If somebody posts possible malicious links, they could get warned about it but still decide to post it anyway (after all, the WOT database could be wrong, but I've only seen this in one case by now and that was a flamewar which sorted itself out). In that case, an automatic flag for moderator attention and review could be issued.