Creating an agent for an Agentic Email bot interface
An agent is a conversational AI-powered assistant that interacts with users to facilitate automated actions and provide context-aware responses. Agents are integrated in applications to enhance user experience by streamlining processes and providing immediate assistance.
Walkthrough
This topic describes how to create an agent that you can use in an Agentic Email bot channel interface.
In the navigation pane of Infinity Studio, click App, and then click Records. Expand the Generative AI section, and then click Agent.
Click Create to open the Create Agent dialog box.
In the Label field, enter a short description or title for your agent. In this example, Address change agent.
In the Scope field, determine whether the agent applies to the application or a selected Case Type.
Agents created in the application scope work at the application level, while agents created for a Case Type are aware of the case context and assist within that specific case. In this example, select Case Type. Select the Case Type that you want to associate with the agent, and then click Create and open.
The agent definition opens on the Definition tab.
In the Agent Instructions field, describe the purpose and responsibilities of the agent.
In the Guardrails field, define the rules and constraints that the agent must follow.
In the Response style and tone field, specify how the agent communicates with users.
Optionally, to provide additional context, in the Additional context section, add a data page that the agent uses when generating responses. You can also apply a data transform to retrieve only specific data from the data page.
To facilitate starting a conversation with the agent, in the Guided suggestions section, add clickable suggestions that highlight what the Agent can help with and support common actions. Optionally, you can provide additional instructions to guide how the agent responds to the question.
On the Knowledge tab, select a Buddy to allow the agent to source additional information. In this example, select pyKnowledgeBuddy.
You can select one of the Buddies provided by default or your own Buddy.
On the Settings tab, in the Model section, select the AI model that the agent uses.
Click Save to save the agent.
You have reached the end of this demonstration.
This Topic is available in the following Module:
If you are having problems with your training, please review the Pega Academy Support FAQs.
Want to help us improve this content?