DNS Harvest isn't anything fancy, it's a DNS dictionary bruteforce tool.
Right now, it's limited to IPv4 and domain-existence reconnaissance. Hopefully soon it'll be expanded =]
./dnsharvest.py -d example.com -i data/small.lst
The input file, -i in the example above, should be a single-column dictionary of words. If they're not valid subdomain names, the lookup will fail gracefully and move on.