Based on the Report and the media available on a ticket, SnapCall can generate next-step recommendations that guide your third-party AI agent through the following stages of automation.
This feature is primarily intended for AI agents to continue automated workflows, but it can also be used by a human support agent to quickly understand what actions should be taken next based on the analyzed context and available media.
1. Define and Create Your Flow
Start by creating a new Flow or editing an existing one. Provide a clear name and description, and include as much detail as possible about the target media SnapCall should look for in the ticket. The more specific this information is, the more accurate and relevant SnapCall’s recommendations will be.
2. Enable “Recommendations”
Navigate to the Outcome tab of your Flow and enable the Recommendation option. When activated, a prompt field will appear, allowing you to add additional context or instructions that SnapCall should consider when generating recommendations.
You can enter a written prompt describing internal procedures or include a link that directs to a specific process or documentation. If you choose to add a link, make sure it is publicly accessible so SnapCall can retrieve the content during analysis.
This setup ensures that your AI agent (or human agent) receives actionable guidance directly from SnapCall, enhancing automation efficiency and overall ticket resolution quality. For AI Agents, you can directly access report from the conversation metadata and use this as a variable.