FAQs

1. Does ClearFeed support indexing the knowledge sources partially? Answer: Yes, knowledge sources can be indexed partially when some URLs/pages from the knowledge source are interested in ClearFeed and can be used for answer generation. You can see the status of the source as Partially Indexed on the KBs page as shown below.

1. I am getting a failure trying to index my web-site. What are the possible reasons? Answer: These failures can be due to different reasons.

   • Unable to access the web-site or URL. This may happen if the web-site is not open to all IP addresses or wrong URL has been entered. Sometimes websites behind a WAF or a service like CloudFlare are not accessible from all crawlers. For such issues - please find the details of our crawler's IP address and user-agent string in the next FAQ entry and white-list them.

   • The web-site may be password-protected. In this case please reach out to ClearFeed support to integrate with the authentication mechanism.

2. What is the User Agent and IP address of ClearFeed's web crawler? Answer: The outbound IP address used by the ClearFeed web crawler is 3.209.240.136. The User Agent is Clearfeed/1.0.

3. Does ClearFeed use data from the knowledge base to train the AI models? Answer: We do not use any customer data, messages, or knowledge sources to train our AI models. Our model is trained on synthetic data, and the feedback provided by users on answers is used only to display metrics. Any future changes to this policy would require explicit opt-ins from customers. Therefore, you can be assured that your data will not be used for model training.