Let's start with the first step of the process to understand what is happening behind the scenes and how to tell if a ticket was modified or automated by Lang.

Once you Zendesk integration is connected, the next step is to setup the integration in order to get the real-time ticket updates and the power of automations.

Inside your project, the Setup—Integrations section shows a button called "Setup Integration". Once the button is clicked, an inactive trigger is created in your Zendesk instance. We also create what is called a webhook to send your tickets to your Lang.ai project. The webhooks can be reviewed in the Admin Center > Apps and Integrations > Webhooks.

By default, this triggers takes into account tickets that come from the Web form and Email channels. It will also only process tickets that are created (e.g. not updated) and that are generated by an end-user. Tickets from internal agents won't be processed. The conditions from this trigger can be modified to fit your needs.

Lastly, there are two options available in the integration: "Sandbox mode" and "Enable tickets tagging". When the sandbox mode is turned on, the trigger that was created is activated. This way, new tickets are sent to your Lang project. When the tickets tagging is activated, we internally enable our API to start updating the Zendesk tickets.

When a ticket is processed by Lang, we include a tag into the ticket to mark it so we make sure we don't process the same automation more than once. You will see tags starting with lang_automated in that case.

You can also tell if a tag was added by Lang by looking into the events log from a ticket. See below the changes that were made:

Lang updates tickets by including new tags, and this change also affects the ticket fields, updating the associated field option in the drop-down.

To understand how drop-down and checkbox fields are related with tags, check this Zendesk article Understanding tags and ticket fields.