The purpose of the RBL is to add common domains (either through base
or strict
RBLs) to a list where you wouldn't expect these domains to make use of generic services in the public domain.
E.g. if you're providing a service where users can add domains, you'd generally want to prevent domains such as google.com or facebook.com to be added as domains - the RBL is to provide the information dynamically and crowdsourced without having to maintain manual lists.
base
: This list should only contain very well known domains
strict
: This list can contain domains that people deem should generally be unavailable for being registered (hosting providers, smaller ESPs, ISPs etc) - applications should ideally have an exception list available for anything that may be contained within the strict
list.
We automatically import domains from the below lists:
- https://github.com/dyne/domain-list (imported into
base
andstrict
)