Explaining AI Agents

Imagine an electronics retailer introduces a large set of new digital cameras across five languages. However, it faces issues like incomplete product data, missing translation details in a few languages, and poor search performance. The retailer uses AI agents in the SAP CX AI Toolkit to reduce the manual effort.

How AI Agents Take Different Roles:

• Agent A reviews all the new camera listings and finds gaps. It identifies missing details such as megapixels, ISO range, and video capabilities. It also detects translation problems on product pages in different languages. The agent flags products that need improvement.
• Agent B uses the improved product data to help customers find the right products and receive accurate product recommendations.
• Agent C supports customers with general inquiries such as store hours, return policies, warranty information, and other service-related questions.

Results:

Together, these agents use curated internal knowledge sources and data to deliver a smooth, effective customer experience.

Rather than building these agents from scratch, the SAP CX AI Toolkit offers a faster path where administrators can leverage pre-built AI agent templates. Let’s explore how AI agents work and how AI agent templates in the SAP CX AI Toolkit help you build agents quickly.

AI agents are configurable, autonomous components in the SAP CX AI Toolkit. They harness data from SAP Customer Experience (CX) products, including SAP Commerce Cloud, SAP Sales Cloud, and SAP Service Cloud, to make decisions and drive business actions. These agents achieve user-defined goals through a cycle of planning, reasoning, self-reflection, and action.

Administrators manage each agent's entire lifecycle directly in the dedicated AI Agents workspace, as shown below.

Key administrative capabilities include:

• Lifecycle Management: Creation, configuration, testing, activation, and deactivation of agents.
• Governance: Access to filtered logs (by date range, errors only, or specific record IDs) for troubleshooting.
• Data Grounding: Integration with relevant data sources to ensure agents operate reliably and produce grounded outcomes.

We will explore this further when discussing individual AI agents.

AI agent templates are prebuilt blueprints in the SAP CX AI Toolkit that accelerate time-to-value by providing ready-to-use configurations for common CX use cases. When administrators create a new agent, they use a wizard that starts with template selection, as shown below.

Currently, there are three specialized built-in AI agent templates available for SAP Commerce Cloud:

• Digital Service Agent: Handles natural language inquiries for automated support.
• Shopping Agent: Acts as a 24/7 digital sales associate for the storefront.
• Catalog Optimization Agent: Manages the quality and consistency of product data.

Now connect these agents to the opening scenario: Agent A maps to the Catalog Optimization Agent, Agent B to the Shopping Agent, and Agent C to the Digital Service Agent.

We’ll cover each of these agents in detail in the following lessons.

In this lesson, you learned:

• AI Agents are autonomous, data-driven components that plan and act to achieve goals using SAP CX data.
• AI Agent Templates are prebuilt agent blueprints that simplify and speed up deployment for common scenarios like digital service and shopping assistance.
• Administrative Controls and reliable data integrations are essential for maintaining safe, grounded, and effective agent operations.

For more information, check the AI Agents page, which provides a comprehensive overview of AI agents, their features, and detailed instructions.