Recently, there appears to be a significant problem with bot accounts posting unsolicited links for the "cheap viewer" scam. While it's possible for individual accounts to block links entirely, this doesn't resolve the issue of the bot account as a whole.

To this end, I believe there should be a ban on any account where their first time chat in a given channel includes any kind of link. This would be an option set by Twitch themselves and not configurable by streamers individually. In addition, it would also be helpful if a system could be put in place where any account that tries to post a link as their first message is automatically reported as a suspected bot account and subject to manual review.

If you wanted to expand on this idea further, you could also potentially have a tiered restriction list. As a rough example:
- 1) First Time Chatter - Automatic channel ban and report to Twitch as outlined above.
- 2) New Chatter, NOT First Time - The first X messages sent in a channel are automatically monitored for any non-Twitch links and if a link is sent, the account is automatically flagged for manual review. Automatic channel bans could also be considered here, but this could be done at broadcaster's discretion.
- 3) Regular Viewer/Chatter - Current link moderation rules apply.

Pros:
- Would reduce the volume of spambot accounts getting messages through on otherwise unprotected accounts.
- Would reduce broadcaster/moderator aggravation for having to remove and report these accounts.
- Ensures a better experience for all legitimate viewers.
- Restricts botter's means to spread their spam/scam messages.

Cons:
- At least to begin with, automatic reports would result in increased workloads for Twitch staff. This would ideally reduce over time though.
- While unlikely, the potential for a small amount of false positives would exist. This is why manual review by Twitch staff rather than automatic processing would be required.
- This idea only works if a valid URL exists in a message, and would not take into account any messages that obfuscate links (i.e. myfakelink dot com); this type of obfuscation would still need to be blocked at the broadcaster/moderator level