SLF 1 - 3

Sunrays Limited (SL) wants to access Salesforce’s generative AI features but has concerns over its company data being exposed to third-party large language models (LLMs). Specifically, SLvwants the following capabilities to be part of Einstein’s generative AI service. - No data is used for LLM training or product improvements by third-party LLMs. - No data is retained outside the SL’s Salesforce org. - The data sent cannot be accessed by the LLM provider. Which property of the Einstein Trust Layer should the AI Specialist highlight to SL that addresses this requirement?. Prompt Defence. Zero Data Retention Policy. Data Masking.

What is the correct process to leverage Prompt Builder in a Salesforce org?. Select the appropriate prompt template type to use, select one of Salesforce’s standard prompts, determine the object to associate the prompt, select a record to validate against, and associate the prompt to an action. Select the appropriate prompt template type to use, develop the prompt within the prompt workspace, select resources to dynamically insert CRM-derived grounding data, pick the model to use, and test and validate the generated responses. Enable the target object for generative prompting, develop the prompt within the prompt workspace, select records to fine tune and ground the response, enable the Trust Layer, and Associate the prompt to an action.

An AI Specialist wants to include data from the response of external service invocation (REST API callout) into the prompt template. How should the AI Specialist meet this requirement?. Convert the JASON to an XML merge field. Use External Service Record merge fields. Use “Add Prompt Instructions” flow element.

Sunrays Limited (SL) has a Legacy system that needs to integrate with Salesforce. SL wishes to create a digest of account action plans using the generative API feature. Which API service should SL use to meet this requirement?. Rest API. Metadata API. Soap API.

The sales team at a hotel resort would like to generate a guest summary about the guests’ interests and provide recommendations based on their activity preferences captured in each guest profile. They want the summary to be available only on the contact record page. Which AI capability should the team use?. Agent Builder. Prompt Builder. Model Builder.

Sunrays Limited (SL) is preparing and defining success criteria for Agentforce Testing Center test cases. Which details should SL specify as the expected output to ensure the tests accurately reflect the agent's functionality?. Expected Topic API Name. Expected Flow API Name. Expected Prompt Template Name.

Sunrays Limited (SL) needs consistent pass/fail logic for agent testing. Which Testing Center capability provides that?. Use customer rating as a proxy for correctness. Run a script on event logs to identify the failed utterances. Use structured batch testing with validation per test utterance.

A Service Agent at Sunrays Limited (SL) is designed to help customers resolve issues by searching against knowledge articles. Knowledge articles have PDF attachments that add critical details. UC reports that the agent provides excellent summaries of the knowledge articles, but seems completely unaware of the PDF attachments. How should an Agentforce Specialist configure the Data Cloud search index to include the content of these attached files?. Increase article chunk size and token limits for Knowledge indexing so larger contexts capture attachment references. Enable 'Include Related Attachments’ for Knowledge__kav and map the ContentDocumentLink unstructured data model object (UDMO). Use Data Cloud's ‘Include Attachments’ option and select the ContentDocumentVersion unstructured data model object (UDMO).

Sunrays Limited (SL) wants to use an Al agent to answer questions about warranties. Warranty information has already been uploaded as unstructured data in Data Cloud. When answering user questions, the results must be filterable by product line and ranked by recent updates. Which approach should the Agentforce Specialist implement?. Use the default retriever which automatically accounts for recency ranking. Build a custom retriever in Einstein Studio with product line filters and recency ranking. Apply semantic embeddings with default metadata filters to achieve the desired result.

After an agent selects a topic, what is an important factor the reasoning engine uses to select the action?. The priority given to each action. The explicit order of actions in the topic. The name and instructions of the actions.

Sunrays Limited (SL) is tracking web activities in Data 360 for a unified contact. It wants to use that information in a prompt template to help extract insights from the data. Assuming that the Contact object is one of the objects associated with the prompt template, what is a valid way for SL to do this?. Call the prompt directly from Data 360 with a web tracking activity included in the prompt definition. Add the activity records as an enrichment related list to the Contact, then pass the Contact into a prompt template workspace using related list grounding. Create a prompt template that takes a list of all Data 360 activity records as input to pass to the large language model (LLM).

A customer wants to analyze complete agent interactions, from the initial user request through to the final resolution, to better understand agent behavior and response quality. Which Agentforce feature should an Agentforce Specialist recommend?. Agent Inspection. Agent Insights. Agent Optimization.

An Agentforce Specialist is creating a prompt template to assist service reps in drafting responses to customer complaints. To ensure the responses are empathetic and helpful, what is a key element to include in the prompt template?. A list of keywords related to customer complaints. A direct instruction to the large language model (LLM) to role-play as a character. The entire history of the customer's previous interactions with the company.

Sunrays Limited (SL) is implementing a Retrieval Augmented Generation (RAG) solution where the knowledge store contains Spanish articles, but users will query the agent in French and English. The Agentforce Specialist needs to ensure the retriever can accurately find relevant articles despite the language differences. What should the specialist do to meet this requirement?. Use the multilingual-e5-large embedding model for French and Spanish language users. Use the multilingual e5-large embedding model for French and English language users. Use the multilingual-e5-large embedding model to handle all languages.

Sunrays Limited (SL) is designing an agent to assist with order management and dealer support automation. The agent must verify a dealer's credentials before granting access to order details. The team has already: - Declared a variable is_verified to track verification status - Configured an action that verifies the dealer's credentials They plan to restrict access to the order details subagent (formerly known as a topic) using a guard condition based on is_verified. What must the team do to ensure the order details subagent becomes available only after a dealer is successfully verified?. Add an available when: @variables.is_verified = true condition to the order details subagent. Declare the is_verified variable as immutable so it cannot be modified during the session. Update the is_verified variable to true after the verification action succeeds using @utils.setVariables.

An Agentforce Specialist at Sunrays Limited (SL) previously recommended removing superseded mortgage policy documents ingested into Data 360 to prevent the agent retrieving outdated terms. Legal has now confirmed that all historical versions must remain in the system simultaneously, as each version remains contractually binding for customers who signed under it. Which revised recommendation should the specialist make?. Configure a custom retriever in AI Models (formerly Einstein Studio) with a dynamic filter on the policy version metadata field, populated at runtime from the customer's Contract record in CRM, ensuring retrieval is deterministically scoped to the document version applicable to that customer's agreement before similarity ranking is applied. Add the policy version identifier as a prepend field in the search index so every chunk is tagged with its applicable version at indexing time, enabling the large language model (LLM) to identify and select the correct version during response generation. Create a separate data stream for each policy version and configure the agent to query the appropriate unstructured data model object (UDMO) corresponding to the customer's contract date, isolating each version's vectors from the others.