SLF 2 - 4

Sunrays Limited (SL) has a Flex prompt template that has been used for the last 3 months to answer questions based on user input. Now, SL wants to give a PDF as a second input. What is the best approach for the Agentforce Specialist to meet this requirement?. Reindex in order to add a new resource to an existing template. Add a Resource anytime by navigating to the Resources section and configuring inputs. Discard the current Flex template and create a new one with a resource.

Sunrays Limited (SL) is deploying a new customer service agent using Agent Script. The Agentforce Specialist needs to update conversational state, navigate between different subagent (formerly known as a topic), and hand off conversations to a human agent when necessary. To minimize technical debt, the Agentforce Specialist wants to implement these capabilities without building any custom backend code. Which statement correctly describes how the Agentforce Specialist can implement these specific requirements?. The Agentforce Specialist can use @utils.setVariables, @utils.transition, and @utils.escalate directly. The Agentforce Specialist can use most utilities directly, but @utils.escalate requires a custom Apex class to define the escalation queue. The Agentforce Specialist can use most utilities out of the box, but @utils.transition requires a flow to define the target subagent at runtime.

Sunrays Limited (SL) is configuring a customer service agent and needs to restrict a specific "Process Refund" action so that it is only visible to the reasoning engine when a customer's account is flagged as active. An Agentforce Specialist is writing the available when clause in Agent Script to evaluate the @variables.IsActive variable. Which operator should the Agentforce Specialist use for this condition?. The Agentforce Specialist must use the operator; Agent Script uses == strictly for comparison in conditionals. Both and are valid and interchangeable; the Agentforce Specialist can use either operator to evaluate conditional logic in Agent Script. The Agentforce Specialist must use the operator; Agent Script uses for both variable assignment and condition comparison.

During retrieval augmented generation (RAG) quality testing, an Agentforce Specialist notices that tabular information from a custom Data 360 Document Ingestion Pipeline is losing its context because the data is scattered across multiple separate chunks. What is the most appropriate approach to resolve this?. Change the search index's parser from the default to Docling. Use an ensemble retriever to stitch multiple chunks back together dynamically. Switch the search index configuration to use only keyword search scoring.

An Agentforce Specialist is trying to troubleshoot a reported issue from an end user and they do not see the session in the Processed Sessions tab of Agentforce Observability. What should the specialist take into consideration?. The agent has not yet been registered in the Observability app so no sessions are processed. The session is queued to be processed and can be viewed on the Unprocessed Sessions tab. Sessions are batched on a 24-hour rolling process and the logs must be checked again the next day.

Sunrays Limited (SL) has built a custom recround AI responses with technical documentation. SL now requires that a summary field from the search index be included in the output to provide a brief overview of each retrieved document. What is the recommended course of action for SL?. Create a completely new custom retriever with an entirely revised configuration that incorporates the summary field, and delete the current retriever. Editing the existing retriever is not possible. Edit the existing retriever version to add the summary field to the list of fields returned by the retriever, save the changes, and activate the new version so that prompt templates use the updated configuration. Use only the default retriever, which will automatically include all available fields from the search index, including the summary field, without any need to create or manage custom retrievers.

An Agentforce Specialist is setting up Salesforce Knowledge as the data source for a Data Library and must select "Identifying Fields" to help Agentforce locate the right information. The Knowledge article contents are rich, detailed, and quite long. What should the Agentforce Specialist use as Identifying Fields?. The main contents of the article to ensure the most relevant article is used. Any text or text area fields that provide a concise summary of the article. Any standard or custom formula field that contains a suitable concatenated key.

Sunrays Limited (SL) has long, complex legal agreements. An Agentforce agent must be able to retrieve specific clauses that are often nested within larger sections. The standard chunking method is failing to capture the full context of these clauses. Which chunking strategy should an administrator use to preserve the structure of the documents?. Implement a smaller chunk size. Implement a larger chunk size. Implement a keyword-based chunking strategy.

An Agentforce Specialist is working declaratively on version 4 of a prompt template which is significantly different than the previous version 3. A bug is then identified in version 3 and requires a hotfix to be deployed to production immediately by another team. The team will deploy this change as a new version. What should the specialist do to ensure that their work on the latest version is not lost and does not conflict with the new version?. Create a new prompt template with the bug fix, update all references to the new version, and delete the old prompt template. Save their work in version 4, and allow the template to be deployed with a new version 5 containing the fix. Copy and paste version 4 into a backup file or version-control alternative, and allow the other team to overwrite the contents of version 4 with the bug fixed version.

Sunrays Limited (SL) has built a Service Agent for its hospitality brand grounded on a knowledge data library containing 200 FAQ articles covering reservations, amenities, and cancellation policies. The project team estimates that manually authoring test cases to achieve adequate knowledge coverage would take 3 weeks. A junior administrator proposes that Testing Center can significantly accelerate this process. Which step should an Agentforce Specialist take to generate a comprehensive initial test suite without manual authoring of individual test cases?. Upload test cases. Generate test cases based on the knowledge available to the agent. Generate test cases based on subagent (formerly known as a topic) and actions.

Universal Containers has deployed Agentforce to handle customer order tracking, returns, and loyalty support. The agent needs to balance conversational flexibility for customer inquiries with guaranteed execution of identity verification steps before accessing the account information. The development team is evaluating how to structure the agent's instruction pattern to meet both requirements. Which statement correctly describes hybrid reasoning in Agentforce Agent Script?. Hybrid reasoning uses multiple large language model (LLM) models simultaneously, with one model handling conversational responses and another model executing deterministic business logic through Flow integration. Hybrid reasoning requires Canvas View for declarative instructions and Script View for procedural instructions, as each editor provides different capabilities for the respective instruction types. Hybrid reasoning combines declarative natural language instructions that allow large language model (LLM) interpretation with procedural instructions using the '->' prefix that enforce guaranteed execution order.

Before activating a custom agent action, an Agentforce Specialist would like to evaluate multiple real world user utterances to ensure the action is being selected appropriately. Which tool should the Agentforce Specialist recommend?. Agentforce Builder. Testing Center. Prompt Builder.

Sunrays Limited (SL) wants to allow its service agents to query the current fulfillment status of an order with natural language. There is an existing autolaunched flow to query the information from Oracle ERP, which is the system of record for the order fulfillment process. How should an Agentforce Specialist apply the power of conversational AI to this use case?. Create a Flex prompt template in Prompt Builder. Create a custom Agent action which calls a flow. Configure the Integration Flow Standard Action in Agent Builder.

What should an AI Specialist consider when using related list merge fields in a prompt template associated with an Account object in Prompt Builder?. The activities related list on the Account object is not supported because it is a polymorphic field. If person accounts have been enabled, merge fields will not be available for the account object. Prompt generation will yield no response when there is no related list associated with an Account in runtime.

An Agentforce Specialist turned on Einstein Generative AI in Setup. Now, the Agentforce Specialist would like to create custom prompt templates in Prompt Builder. However, they cannot access Prompt Builder in the Setup menu. What is causing the problem?. The Prompt Template User permission set was not assigned correctly. The Prompt Template Manager permission set was not assigned correctly. The large language model (LLM) was not configured correctly in Data Cloud.