Resolver employs a variety of techniques to protect your chatbot from spam and abuse.
Honeypot fields
There are two types of honeypot fields used in Resolver.
The first is a field that should be left empty. Simple bots will mistakenly enter info into this field, which will be detected and cause the submission to be discarded.
There is another field that requires JavaScript to be rendered and must be filled out. Many bots carrying out spam submissions do not run JavaScript, so the field is not included in their submission, and it is discarded.
Rate limiting
There is rate limiting in place to prevent anyone from sending messages too quickly. This prevents a bot from spamming messages rapidly. The rate limiting is implemented in numerous ways to slow down any bad actors.
Character limit
There is a character limit that prevents anyone from submitting a large block of text that would eat up tokens.
Relevancy targeting
The pre-written instructions teach the chatbot not to reply to unrelated questions. This prevents anyone from using your chatbot for general Q&A unrelated to your business.
Additional strategies
We use and highly recommend Cloudflare for additional bot protection. The robust firewall and DDoS protection offered in their free plan can stop bots from accessing your site in the first place, which saves bandwidth and hosting costs.
You can also use the free Wordfence plugin to add a firewall that will block bots from attempting spam submissions.