WorksheetsSet1 85% New Pass - part 1
Total questions: 30
Worksheet time: 15mins
Universal Containers has a requirement to provide a sales summary for its sales reps who are using Employee Agents, but they are not happy with the default answer. Which best practice should the Agentforce Specialist recommend?
Update the standard record summary action.
Create a Record Summary custom prompt template.
Create a Knowledge Answer custom prompt template.
Universal Containers (UC) recently attended a major trade show and received thousands of new leads from event badge scans. UC 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, and free up reps to focus on building relationships and closing deals. Which Agentforce capability should UC implement to meet these goals?
Commerce Agent
Sales Coach
SDR Agent
An Agentforce Specialist is creating a custom agent action. The topic is selected correctly, but the action is not. Which setting should the Agentforce Specialist test and iterate on to ensure the action performs as expected?
A. Action Instructions
B. Classification Description
C. Action Scope
Universal Containers' service team wants to customize the standard case summary response from Agentforce. What should the Agentforce Specialist do to achieve this?
A. Customize the standard Record Summary template for the Case object.
B. Summarize the Case with a standard agent action.
C. Create a custom Record Summary prompt template for the Case object.
Universal Containers (UC) is considering using a Field Generation prompt template type. What should UC check before creating the Field Generation prompt to ensure it is possible for the field to be enabled for generative AI?
That the org is set to API version 59 or higher
That the Lightning page layout where the field will reside has been upgraded to Dynamic Forms
That the field chosen is a rich text field with 255 characters or more
Universal Containers wants to systematically validate agent responses before deployment using a scalable testing process. Which Testing Center approach should the company implement?
A. Use pilot users in production to flag incorrect responses post launch.
B. Manually interact with the agent in Builder until responses seem correct.
C. Upload a structured CSV test template and run batch test cases in Testing Center.
Universal Containers (UC) needs to create a custom prompt template that can be called from a Lightning web component. Which prompt template type should UC create?
Flex
Field Generation
Sales Email
An Agentforce Specialist wants to troubleshoot an agent that is hallucinating weblinks. The agent has an action that uses a prompt template, which is using a knowledge retriever, to generate the output text that the agent will use. Which process is appropriate to find the root cause of the hallucination behavior?
Examine the topic name and classification description for hallucination guardrails.
Examine the prompt instructions and contents of the chunks shown in the resolved prompt output.
Examine the topic instructions and ensure the word "ALWAYS" is used in the hallucination guardrails.
Universal Containers wants to keep retrieval accurate as product documentation changes frequently. Which approach should the company implement?
Manually delete the stale data chunks.
Rebuild the search index.
Leave embeddings unchanged even if content is updated.
An Agentforce Specialist needs to create a prompt template that extracts the customer's name, phone number, and case number from a block of text, and nothing else. How should the Agentforce Specialist structure the prompt to ensure the large language model (LLM) doesn't include extra conversation or text?
Ask the LLM to extract and only output the important information in the text.
Ensure in the prompt that the LLM has been told to only use name value pairs in the response.
Use well-defined output instructions and provide desired output examples.
Universal Containers' administrator has developed a new agent in a sandbox environment and now wants to deploy it to production. What should the administrator do to deploy an agent?
Export agent components as JSON files and manually import them into production using the Metadata API.
Manually recreate the agent configuration, topics, and actions in production because change sets cannot be used.
Create an outbound change set with all the necessary agent components, then upload to production.
Universal Containers (UC) has a new AI project. What should UC consider when adding a related list on the Account object to be used in the prompt template?
After selecting a related list from the Account, use the field picker to choose merge fields in Prompt Builder.
Prompt Builder must be used to assign the fields from the related list as a JSON format.
The fields for the related list are based on the default page layout of the Account for the current user.
An administrator at Universal Containers has successfully deployed a new agent from a sandbox to production using a change set. The agent uses a prompt template that invokes a Salesforce flow to perform a complex calculation. In production, when users interact with the agent, it fails with an error message every time the flow is supposed to run. The flow was included in the change set and is present in production. What is the most likely cause of this issue?
The change set did not include the dependent Apex classes for the flow.
The user in production does not have permission to run the flow.
The flow was not manually activated in the production org after the deployment.
An Agentforce Specialist is building a Flex prompt template. Which best practice should the Agentforce Specialist follow when creating the Flex prompt template?
Provide the large language model (LLM) with contextual information and give it a role such as a sales or support rep.
Ground the large language model (LLM) with a retriever and create a custom field to store the LLM-generated response.
Ground the large language model (LLM) with account data and create a custom field account summary to store the LLM-generated response.
Universal Containers deploys a new Agentforce Service Agent into the company's website but is getting feedback that the Service Agent is not providing answers to customer questions that are found in the company's Salesforce Knowledge articles. What is the likely issue?
The Agentforce Service Agent user is not assigned the correct Agent Type License.
The Agentforce Service Agent user was not given the Allow View Knowledge permission set.
The Agentforce Service Agent user needs to be created under the standard Agent Knowledge profile.
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
Prompt Builder
Testing Center
What is automatically created when a search index is created in Data Cloud?
A. A dynamic retriever to allow runtime selection of retriever parameters without manual configuration
B. A predefined Apex retriever class that can be edited by a developer to meet specific needs
C. A default retriever that shares the name of the custom search index
What does it mean when a prompt template version is described as immutable?
Only the latest version of a template can be activated.
Every modification on a template will be saved as a new version automatically.
After a prompt template version is activated, no further changes can be saved to that version.
Coral Cloud Resorts (CCR) sees the agent forgot the dietary/activity preferences gathered earlier. They need those preferences to persist throughout the session. What should CCR implement?
Create a context variable to capture/store customer preferences as action outputs.
Rely on natural conversation memory and instruct the agent to look back.
Configure custom variables to capture/store customer preferences from action outputs.
Coral Cloud Resorts is implementing Agentforce retrieval. Customers sometimes type ambiguous terms (for example, “package” could mean vacation package or baggage). Which retrieval strategy best balances precision and contextual disambiguation?
Use hybrid search, which combines keyword matching for precision with semantic embeddings for context.
Use keyword search only, which prioritizes exact term matching but risks missing contextual meaning.
Use semantic search only, which captures intent but may struggle with ambiguous terms when no context is provided.
Coral Cloud Resorts is about to start testing its concierge agent with guests. Which metrics should be captured to monitor the performance, correctness, and user experience?
Response performance, tone, and CSATs
Response times, accuracy and relevance of answers, and resolution success
Agent performance, token usage, and conversation duration
An Agentforce Specialist at Universal Containers (UC) is building with no-code tools only. They have many small accounts that are only touched periodically by a specialized sales team, and UC wants to maximize the sales operations team's time. UC wants to help prep the sales team for calls by summarizing past purchases, displaying products the contact has shown interest in (with data captured via Data Cloud), and providing a recap of past email and phone conversations that have transcripts. Which approach should the Agentforce Specialist recommend to achieve this goal?
Deploy UC's own custom foundational model on this data first.
Use a prompt template grounded on CRM and Data Cloud data using standard foundation models.
Fine-tune the standard foundational model due to the complexity of the data.
Universal Containers has a custom agent action calling a flow to retrieve the real-time status of an order from the order fulfillment system. For the given flow, what should the Agentforce Specialist consider about the running user's data access?
The custom action adheres to the permissions, field-level security, and sharing settings configured in the flow.
The agent will always run flows in system mode so the running user's data access will not affect the data returned.
The flow must have the 'with sharing' permission selected in the advanced settings for the permissions, field-level security, and sharing settings to be respected.
Universal Containers has created an Employee Agent. Which step should an Agentforce Specialist take to connect the agent with a Slack channel?
Create an embedded service deployment and connection between Salesforce and the Slack workspace.
Create an Omni Channel flow and connection between Salesforce and the Slack workspace.
Create a connection between Salesforce and the Slack workspace.
When is the Agent-to-Agent (A2A) protocol an appropriate communication choice?
When agents need to invoke third party API
When agents need to collaborate
When agents need to access tools
Coral Cloud Resorts wants to cover a broad range of user phrasing when testing its FAQ agent. Which Testing Center feature meets that need?
Uploading only a small set of manually written prompts
Relying on live customer logs to capture phrasing diversity after deployment
AI-generated synthetic test utterances based on natural language variations
Universal Containers (UC) wants to empower its marketing team with AI capabilities that help employees quickly find campaign data, generate creative content, and manage project tasks. The solution should also allow marketers to receive personalized support, surface relevant information, and complete work directly in Salesforce.
Employee Agent
Service Agent
Sales Coach Agent
Universal Containers has PDF maintenance guides in an external folder, not yet in Salesforce. The team wants a standard, clicks-only setup for the Service Agent to use these documents. Which approach should the Agentforce Specialist implement?
Configure Data Cloud to ingest file attachments and create custom index and retriever for product record and attachment data.
Upload the PDFs as File source in the Agentforce Data Library which will build a Search Index, and create a retriever to ground responses from those documents.
Paste external PDF links into topic instructions and rely on the model to follow them, avoiding configuration of a retrieval source, index, or retriever action.
A company wants to retrieve patient history details to augment the AI agent response. The company wants to leverage the Data Cloud search index feature. What is best practice when considering retrieval-augmented generation (RAG) for information that may contain personally identifiable information (PII)?
Depend on the agent’s prompt to avoid exposing PII.
Encrypt embeddings, but still index PII records.
Mask sensitive fields and index only non-PII data.
A developer is using the Salesforce CLI to deploy agent components from a sandbox to production. They recently made a change to several topics, instructions, and actions. Which metadata component should the developer include in their package.xml file that contains all of the topics and actions an agent will interact with?
EinsteinAiPlannerBundle
genAiPlannerBundle
BotBundle
