Skip to main content

How does SnapCall returns AI output?

Written by Alessandro Angioni
Updated yesterday

When you create a Flow and add AI skills into it, you can also define how the output should be added into the ticket or conversation. Here below you find a list of channels where you can add AI output.

Ticket comment

Can be public or private, and you can define the type when editing your Flow, on the 'Outcome' tab. The output can be added as Plain text, bullet points, JSON, or any other format

Custom field

We support any type of custom fields in Zendesk, however you need to make sure the value extracted is compatible with the field type you want to update, otherwise Zendesk won't allow to store the information.

Tags

SnapCall automatically adds tags based on the Flow used and the features enabled within it. For example, when a Flow is applied to a ticket, a tag with the Flow name is added on it, along with tags that indicate whether an image is AI-generated or sourced from the web (stock).

Metadata

If the Flow is linked to a conversation, SnapCall can also send AI output into its metadata. If you're using Advanced AI agent, you can extract output and save them as a variables using the following keyword

  • Content recognition is saved as 'snapcall_description_1', where the number at the end indicates the step where the AI skill was applied into.

  • Summary is saved as 'snapcall_summary'

  • Transcript is saved as 'snapcall_transcript'

  • Any output stored in custom fields will be saved as 'zen:ticket_field:31012753155346', where 31012753155346 is dynamic and represent the id of the custom field selected on your SnapCall Flow to store the output.


​

Did this answer your question?