Knowledge Base Data Sources
Knowledge Bases in Flow XO are a easy to use and powerful way to automate the process of providing complex questions about your business. As simply as Knowledge Bases are to use, if you have a lot of content, it can be somewhat cumbersome to load all of your documents into your knowledge bases manually. However, with Data Sources, as long as your content is publicly available on the web, such as pages on your website or help center, Flow XO can automatically keep your online content synchronized with your Flow XO knowledge base so that it's always up to date.
Knowledge Base data sources work by reading your web content every 24 hours and creating a text-based version of each web page as a separate document in your knowledge base, which can then be searched and fed into an AI language model to automatically answer your users questions.
NOTE: Knowledge Base data sources are in technical preview, and only available to paid Flow XO accounts. There are no limits on the number of pages that can be synchronized during the preview, but we may add limits or additional pricing for very large websites in the future.
To synchronize your web content with your Knowledge Base, you must first create a Knowledge Base that will act as the repository of your web content. Once you have created a Knowledge Base, you can click the 'New Data Source' button on the homepage of your Knowledge Base to configure the synchronization job.
Configuring the sync is simple. Just add a name, an optional description, and then list each starting URL on a separate line in the Start URLs section. You do not need to list every page of your website, only the home page. Our web crawler will then scrape your site to discover all your content. If there are sections of your website that you do NOT want included in your Knowledge Base, you can add the URL to those sections in the URL Prefixes to Exclude section. URLs that our web crawler discover that begin with one of the URLs in the URL Prefixes to Exclude will NOT be read by our import mechanism.
In the example above, we do not want URLs that start with flowxo.com/share or flowxo.com/app to be included. So our import will ignore URLs like https://flowxo.com/share/123xyz, but will include urls such as https://flowxo.com/articles/1234
Once you have configured your sync, click 'Save Changes'
NOTE: At this time, once you create a sync job, you cannot edit the settings. If you need to change the Start URLs or the URL Prefixes to Exclude later, you will need to delete the sync job and create a new one with the corrected settings. Doing this will not remove any of your existing imported documents, so it is safe to do at any time.
After you save your changes, we will immediately begin reading your site. You should start seeing your web-pages show up as knowledge base documents within a minute or two, and, depending on the size of your site, may take anywhere from a few minutes to an hour to fully import your content. The sync job will run every 24 hours to gather added or updated content until you disable or delete the data source.
And that's all there is to it! Let us know if you have any questions or feedback, or if you have any trouble with this feature. As it is still very new, there may be issues, so please let us know right away at firstname.lastname@example.org if something doesn't look right.