We often ask to post the output of network-related commands to diagnose and resolve a problem. Even after the problem is resolved, all this information remains available to any Internet user and can also be indexed by public and private search engines.
The output of some commands may contain sensible IDs and IP addresses, allowing the local or remote fingerprinting of the user's host: an unexperienced user is an easy target for crackers who may compare the question content with the result of a network fingerprinting (for example an Nmap active scan). This is the first step for social engineering and vulnerability exploiting.
Since an user may not know which information is dangerous to share, I propose to add an option to flag explicitly every question that contains sensible data and that (if possible) those questions should be hidden until the sensible information is removed.