SLF 2 - 3

Sunrays Limited (SL) implements a custom retriever to improve the accuracy of AI-generated responses. SL notices that the retriever is returning too many irrelevant results, making the responses less helpful. What should SL do to ensure only relevant data is retrieved?. Define filters to narrow the search results based on specific conditions. Increase the maximum number or results returned to capture broader datasets. Change the search index to a different data model object (DMO).

Sunrays Limited (SL)'s Agent Action includes several Apex classes for the new Agentforce Agent. What is an important consideration when deploying Apex that is invoked by an Agent Action?. Apex classes invoked by an Agent Action may be deployed with less than 75% test coverage as long as the agent is not activated in production. The Apex classes may bypass the 75% code coverage requirement as long as they are only used by the agent. The Apex classes must have at least 75% code coverage from unit tests, and all dependencies must be in the deployment package.

Sunrays Limited (SL) configured a new PDF file ingestion in Data Cloud with all the required fields and created the mapping and the search index. SL is now setting up the retriever and has noticed that a required field is missing. How should SL resolve this?. Update the search index to include the desired field. Modify the retriever's configuration to include the desired field. Create a new custom Data Cloud object that includes the desired field.

An Agentforce Service Agent, who has been successfully assisting customers with service requests in Salesforce, is now unable to help customers with issues related to a new product replacement process. The company recently implemented a custom Product Replacement object in Salesforce to track and manage these replacements. Which Agentforce Agent User change must be implemented to address the issue?. The permission set assigned to the Agent User needs Read access to the custom Product Replacement object. The profile assigned to the Agentforce Agent User needs. Al training permission to the custom Product Replacement object. The permission set group assigned to the Agent User needs to grant access to the Product Replacement flow.

Sunrays Limited (SL) users are complaining that agent answers are not satisfactory. The agent is using PDF files as a knowledge source. How should SL troubleshoot this issue?. Analyze the data mapping between source fields and Data Cloud object fields. Verify the retriever's filter criteria and data source connection. Check that the agent has the PDF the field permission access for the data library.

Which scenario best demonstrates when an agentforce data library is most useful for improving an Al agent's response accuracy?. When the Al agent needs to combine data from disparate sources based on mutually common data, such as customer Id and product Id for grounding. When the Al agent must provide answers based on a curated set of policy documents that are stored, regularly updated, and indexed in the data library. When data is being retrieved from snowflake using zero-copy for vectorization and retrieval.

Sunrays Limited (SL) is setting up the data library configuration within the Agentlorce Builder. What is true regarding Agentfarce Data Libraries?. An agent can have only one data library assigned to it. Each data category can only have one data library. Only data library owners can assign it to the agent.

Sunrays Limited (SL) needs to capture and store detailed interaction data for all agents. Which feature should help SL get a full view of the agent's behavior from start to finish, including reasoning engine executions, actions, prompt and gateway inputs/outputs, error messages and final responses?. Utterance Analysis. Agentforce Session Tracing. Agentforce Analysis.

Sunrays Limited (SL) wants to keep the retrieval accurate as product documentation changes frequently. Which approach should the company implement?. Rebuild the search index. Manually delete the stale data chunks. Leave embeddings unchanged even if content is updated.

Sunrays Limited (SL) stores case details and updates in several custom fields and custom objects related to the case. SL would like its Agentforce Service Agent to be able to provide information in these fields and related records as part of an answer back to its customers when the customer is asking for updates. Which best practice should SL follow to grant access to this information for the Agentforce Service Agent?. Update the Object and Field access in the AgentforceServiceAgentUserPsg permission set group that is already assigned to the Agentforce Service Agent user. Create a new permission set with the Einstein Agent License and enable Read access to the custom fields and custom objects, and assign it to the Agentforce Service Agent user. Update the Object and Field access in the Einstein Agent User Profile so that the Agentforce Service Agents will always get the necessary access.

Sunrays Limited (SL) recently attended a major trade show and received thousands of new leads from event badge scans. SL is struggling to follow up with each lead in a timely, personalized way. Leadership wants to: -Qualify and nurture leads 24/7. -Provide accurate answers to prospect questions. -Automatically book meetings with qualified prospects. -Free up reps to focus on building relationships and closing deals. Which Agentforce capability should SL implement to meet these goals?. SDR Agent. Sales Coach. Commerce Agent.

Sunrays Limited (SL) has been building an agent using Canvas. It has created an "Issue Refund" action. Business rules dictate that refunds can only be issued if the customer's Account_Tier is "Platinum". Currently, the agent relies on its system instructions to check the tier, but it occasionally hallucinates authorization and attempts to call the refund action for Standard customers, resulting in backend errors. How should the Agentforce Specialist use filters to deterministically ensure the agent cannot process the "Issue Refund" action unless the customer meets the criteria?. Update the agent's system instructions with a strict rule to manually filter out any refund requests if the customer's account tier is not Platinum. Apply a text filter to the Agent Session that automatically blocks the word "refund" from the user's prompt unless they are a Platinum customer. Configure an Action Condition filter on the "Issue Refund" action so that it is only available to the agent when Account_Tier equals "Platinum".

Sunrays Limited (SL) uses a third-party agent for research and an Agentforce agent for customer service. Which purpose-built protocol allows cross-vendor agents to communicate?. Model Context Protocol (MCP). Application Programming Interface (API). Agent-to-Agent (A2A).

Sunrays Limited (SL) is deploying several prompt templates to assist its support agents using Salesforce's standard foundation models. Leadership requires the generated responses to consistently reflect an empathetic and highly professional tone. SL only permits the use of standard foundational large language models (LLMs). What is the most effective prompt engineering technique the Agentforce Specialist should implement in Prompt Builder to fulfill this requirement?. Configure the prompt template tone with a dataset of past interactions using different writing styles, intensifiers, and punctuation to permanently alter the LLM default tone. Include a direct instruction asking the LLM to role-play as a specific character (for example, "Act as an empathetic customer support agent") to provide context and establish the tone. Include multiple-choice picklist questions within the prompt template to systematically test and correct the LLM's understanding of the desired context before generating the output.

Sunrays Limited (SL) is deploying two agents simultaneously: an internal Sales Productivity Agent for employees and a customer facing Service Agent on its Experience Cloud site. An Agentforce Specialist is configuring permissions and needs to understand the correct security context for each. Which statement accurately describes the execution model?. The customer-facing Service Agent inherits the security context of the Experience Cloud site's guest user profile by default, requiring guest profile object permissions to be configured for all actions the agent executes. The internal Sales Productivity Agent executes actions using the authenticated Salesforce employee's own permissions, while the customer-facing Service Agent executes actions as a dedicated Einstein Service Agent User with its own permission set. Both the internal Sales Productivity Agent and the customer-facing Service Agent execute actions as a dedicated Einstein Service Agent User with its own permission set that is configured as required for the agent.

An Agentforce Specialist deployed a Service Agent to an Experience Cloud site and enabled Credential Based User Verification. The specialist notices that all Data Manipulation Language (DML) record updates are showing the "LastModifiedUser" as the authenticated Community User instead of the Agent User. What should the specialist explain to the business about the effect on audit fields?. Credential-Based User Verification has been enabled which in turn respects all of the sharing and field level security. The Flow execution mode for the agent is set 'System Context Without Sharing'. Token-Based User Verification has been enabled which in turn respects all of the sharing and field level security.