Muting words is supported by some clients, but not all. It's also a temporary band-aid. It works on this particular spam attack, but is not likely to work on the next one.

We'll have to determine what word will work for the next one when it happens.

Or we could just use better relays...