Free AGENTFORCE-SPECIALIST Practice Test

Model Builder, a component of the Einstein 1 Studio, is the designated tool for integrating externally-hosted AI models with the Salesforce Platform. It provides a unified interface to connect to and manage various large language models (LLMs), including those hosted on third-party services like Amazon Web Services (AWS SageMaker and Bedrock), Google Vertex AI, or custom models accessible via an OpenAPI specification. This "Bring Your Own Model" (BYOM) capability allows organizations to leverage their existing investments in external data science infrastructure while using the models securely within Salesforce for generative AI applications.

B. App Builder: This is a declarative tool used for creating and customizing user interfaces, such as Lightning pages and applications. It does not handle the integration of external AI models.
C. Copilot Builder: This tool is used to create custom AI assistants and actions. It consumes AI models that have already been connected and registered on the platform via Model Builder.

1. Salesforce Help Documentation, "Model Builder": "Model Builder lets you connect a large language model (LLM) that’s hosted outside of Salesforce... You can connect to models hosted on Amazon SageMaker, Amazon Bedrock, Google Vertex AI, or through an OpenAPI spec." (Salesforce Help, Einstein Generative AI, Model Builder section).

2. Salesforce Developers Blog, "Unlocking AI Potential with Einstein 1 Studio": "Model Builder: A tool that allows you to choose from a variety of LLMs from Salesforce’s trusted partners, or bring your own model (BYOM) and host it on the Salesforce Platform. This flexibility allows you to use the best model for your specific needs." (Einstein 1 Studio Overview).

3. Salesforce AI Documentation, "Einstein 1 Studio Guide": This guide details the components of Einstein 1 Studio, specifying that Model Builder is the foundational tool for making both Salesforce-provided and third-party models available to other tools like Prompt Builder and Copilot Builder. (Chapter: "Integrating AI Models").

Standard Service AI Grounding can include only text-based CRM data. The grounding engine ingests fields whose underlying type is String—plain Text, Long Text Area, or Rich Text Area—to build the context sent to the large-language model. Numeric, pick-list, date, or other non-string types are ignored. Therefore, statement B is correct. Grounding is not restricted to only Case and Knowledge objects, and it executes in the requesting user’s context, so it obeys object and field-level security rather than running in unrestricted system mode.

A. Grounding can reference any object whose string-based fields are selected (e.g., Contact, Account, custom objects); it is not limited to Case and Knowledge.
C. Grounding honors the caller’s sharing and field-level security (user mode); it does not bypass visibility by running in system mode.

1. Salesforce Help, “Configure Standard Service AI Grounding,” Winter ’24, section Supported Field Types (“Only text and rich-text area (String) fields are supported”).

2. Einstein for Service Implementation Guide, Winter ’24, p. 34 – 35 (“Grounding respects org-wide security and runs in user context”).

3. Salesforce Developer Documentation, API v59.0 “Grounding Data Model,” para. 2 (“Picklists, numbers, and dates aren’t evaluated; only string datatypes are ingested”).

When configuring prompt templates in Prompt Builder, it is crucial to understand the limitations of related list merge fields. The Activities related list, which comprises Task and Event records, cannot be used. This is because the WhoId (Name) and WhatId (Related To) fields on these objects are polymorphic, meaning they can link to multiple different types of objects (e.g., a Contact, an Account, an Opportunity). Prompt Builder's merge field functionality does not support this type of complex, polymorphic relationship, and therefore, the Activities related list is not available for use in prompt templates.

B. If person accounts have been enabled, merge fields will not be available for the Account object. This is incorrect. Prompt Builder supports Person Accounts. Merge fields remain available for the Account object, including fields specific to Person Accounts.
C. Prompt generation will yield no response when there is no related list associated with an Account in runtime. This is incorrect. If a specified related list has no records for a given Account, the prompt is still generated, but the section for that related list will simply be empty.

1. Salesforce Help Documentation

"Prompt Template Considerations":

Under the section "Related Lists

" the documentation explicitly states

"Polymorphic relationship fields aren't supported. For example

you can't use the Name field on the Task object (WhoId) or the Related To field on the Event object (WhatId)." This directly supports the correct answer (A).

The same document clarifies the behavior for empty related lists: "If a record has no related list records

the related list section in the prompt is empty." This directly refutes option C.

The documentation also notes support for Person Accounts under the "General" considerations

which invalidates option B.

To ground a prompt template with diverse and dynamic data such as related records (Opportunity Products), external information (Events), and specific examples (Tone), the most effective method is to use a Flow. A prompt-initiated autolaunched Flow can be configured to execute complex logic, including querying related Salesforce records, making callouts to external APIs to fetch event data, and retrieving text from other objects or metadata. The outputs of this Flow are then passed directly into the prompt template as grounding data, ensuring the generated Sales Email is contextual, accurate, and adheres to the desired tone.

B. Create a flex template that takes the records in question as inputs. A prompt template (whether "flex" or standard) is the consumer of the data, not the mechanism that obtains it. This option fails to describe how the data is fetched.
C. Utilize a standard email template and manually insert the required data fields. This describes a manual, non-AI process. It contradicts the question's premise of using a custom prompt template for AI-powered email generation and grounding. ---

1. Salesforce Help Documentation

"Ground Your Prompts with Data Using a Flow."

Section: "Ground Your Prompts with Data Using a Flow"

Content: This document explicitly states

"To ground a prompt template in a Flow

create a prompt-initiated autolaunched flow... Your flow can search for records

summarize data

run business logic

and more. The flow’s outputs are then fed into your prompt template." This directly validates that a Flow is the correct tool to fetch and provide the required data.

2. Salesforce Help Documentation

"Prompt Template Resources."

Section: "Resource Types"

Content: This page lists "Flow" as a primary resource type for prompt templates. It specifies

"Use a flow to run complex business logic and return the output to the prompt template." This confirms that using a Flow is the officially sanctioned method for complex data retrieval for grounding purposes

as described in the scenario.

Copilot Builder is the primary interface for creating, configuring, and managing Einstein Copilot actions. Within the Copilot Builder, there is a specific feature designed to test how the copilot's reasoning engine maps user inputs (utterances) to the correct action. This allows a specialist to enter multiple real-world phrases and verify that the intended action is selected before it is activated. This pre-activation testing is essential to ensure the action behaves as expected and provides a reliable user experience.

A. Model Playground: This tool is used for experimenting with the underlying Large Language Model's (LLM) generative capabilities with different prompts, not for testing the specific invocation logic of a configured copilot action.
B. Agent: An Agent refers to the end-user (e.g., a service agent) who interacts with the copilot, or the copilot persona itself. It is not a development or testing tool. ---

1. Salesforce Help Documentation, "Test Your Einstein Copilot Actions":

Reference: "From Copilot Builder, you can test your copilot actions to see how they respond to different user inputs. Testing helps you refine your action’s instructions and ensure that your copilot selects the right action at the right time." This directly confirms that testing with user inputs is a function of the Copilot Builder.

2. Salesforce Help Documentation, "Create Custom Actions for Einstein Copilot":

Reference: Under the section "Test Your Action," the documentation states: "After you save your action, test it in Copilot Builder to see how it responds to different user inputs... In the Test Action section, enter a phrase that you expect a user to say to invoke this action." This explicitly places the testing of utterances within the Copilot Builder environment.

Salesforce uses a granular permission model for Prompt Builder to adhere to the principle of least privilege. The Prompt Template Manager permission set is specifically designed for users who need administrative control, which includes the ability to create, edit, and manage prompt templates. The Prompt Template User permission set is assigned to end-users who only need the capability to select and execute existing, approved prompt templates in their workflow. This clear separation of duties ensures that template creation is controlled while execution is broadly available to the intended users.

A. The 'Data Cloud Admin' permission set is incorrect for executing templates. It provides broad administrative access to Data Cloud, which is excessive and not the specific permission required for this user-level task.
C. The 'Data Cloud Admin' permission set is not the designated permission for creating prompt templates. The 'Prompt Template Manager' permission set is the correct and most specific permission for this administrative function.

1. Salesforce Help & Training

Give Users Access to Prompt Builder: This official documentation explicitly states the roles of the two primary permission sets.

Section: Give Users Access to Prompt Builder: "To let users create

edit

and manage prompt templates

assign them the Prompt Template Manager permission set."

Section: Give Users Access to Prompt Builder: "To let users run prompt templates

assign them the Prompt Template User permission set."

2. Salesforce Help & Training

Prompt Builder: This overview document reinforces the feature's reliance on specific permissions for governance.

Section: Prompt Builder Considerations: The document outlines the security model

implicitly supporting the use of dedicated permission sets ('Prompt Template Manager' and 'Prompt Template User') for managing and using prompt templates

rather than general administrative permissions like 'Data Cloud Admin'.

Service Replies is a generative AI feature within Salesforce Service Cloud designed specifically to enhance agent productivity. It analyzes the ongoing chat conversation and relevant knowledge articles to automatically draft contextual, ready-to-use responses. Agents can then review, edit if necessary, and send these replies, significantly reducing the time and effort required to respond to customer inquiries. This directly addresses Universal Containers' need to improve productivity in replying to chats.

A. Case Summaries: This feature uses generative AI to condense long case histories into a brief summary, helping agents quickly grasp context, but it does not generate replies for the agent to send.
C. Case Escalation: This is a standard workflow process for routing a case to a higher tier of support based on predefined rules; it is not a generative AI feature for drafting responses.

1. Salesforce Help, "Generate Service Replies for Chat and Messaging": "Save your agents time and effort with Einstein-powered Service Replies. When an agent is active in an Enhanced Chat or Messaging session, they can see AI-generated replies based on the conversation. The agent can then revise the reply and post it in the conversation." This document explicitly states the feature's purpose is to save agent time in chat sessions.

2. Salesforce Help, "Einstein Generative AI for Service": This overview document lists "Service Replies" as a key capability, describing its function as: "Draft service replies for agents in the console for chat, email, and messaging conversations." This confirms its role in generating responses.

3. Salesforce Help, "Summarize Cases and Conversations": This document details the functionality of Case Summaries: "Help your agents and supervisors quickly understand the context of a case or conversation with AI-generated summaries." This clarifies that its purpose is context comprehension, not response generation, differentiating it from Service Replies.

Prompt Builder is a generative AI tool within the Salesforce Einstein 1 Platform specifically designed to create, test, and manage prompts that instruct Large Language Models (LLMs) to generate text-based content. The use case of creating a draft for a support bulletin is a direct application of generative AI, where a pre-defined prompt template can pull in specific data (e.g., patch notes from a custom object) to generate a well-formatted, context-aware text output. This aligns perfectly with Prompt Builder's core function of automating content creation.

B. Creating an Al-generated customer support agent performance score: This is a predictive analytics task that involves calculating a numeric value based on various metrics. This is a use case for Einstein Prediction Builder, not the text-generating Prompt Builder.
C. Estimating support ticket volume based on historical data and seasonal trends: This is a forecasting task, a subset of predictive analytics. This is best handled by tools like Einstein Prediction Builder or Einstein Discovery, which are designed for numerical prediction.

1. Salesforce Help, "Prompt Builder": "Prompt Builder is a tool that helps you create, test, and manage prompts for generative AI... Use prompt templates to create reusable prompts that your users can access to generate content like sales emails, field service work summaries, or agent chat replies." This documentation explicitly states that Prompt Builder's purpose is content generation, directly supporting option A.

2. Salesforce Help, "Einstein Prediction Builder": "Einstein Prediction Builder is a declarative tool that lets you build custom predictions on your Salesforce data with just a few clicks... Predict a numeric value, like the predicted revenue from a deal, or a binary value, like whether a customer is likely to churn." This source confirms that predicting a numeric value (like a performance score in option B or ticket volume in option C) is the function of Prediction Builder, not Prompt Builder.

3. Salesforce Developers, "Prompt Builder Basics" Trailhead Module: "With Prompt Builder, you can create prompt templates that are grounded with your CRM data... For example, you can create a prompt template that generates a personalized email to a customer who has an open case." This reinforces that Prompt Builder's primary use case is generating grounded, text-based content.

The large language model (LLM) within Agentforce operates on a dynamic "plan-and-execute" model. When the sales manager makes a request, the LLM first interprets the natural language to understand the user's intent ("show opportunities") and entities ("my," "open"). It then generates a multi-step plan by selecting the most appropriate, available actions (such as a standard action, Flow, or Apex class) designed to retrieve the specified data. Finally, the system executes this plan, which involves running the selected action to query the Salesforce database for the user's open opportunities and then presenting the results.

B: This describes a traditional, rule-based system, not an LLM. Agentforce uses an LLM for dynamic interpretation and planning, rather than relying on a static set of predefined rules. C: While the process involves matching queries to actions, the term "generates a plan" (in option A) more accurately reflects the LLM's advanced capability to reason and sequence multiple actions if needed, which is more sophisticated than simply matching to a single, predefined "dialog pattern."

1. Salesforce Help

Einstein Copilot: How It Works: "When a user interacts with Einstein Copilot

the copilot orchestrates a response by reasoning over the user’s prompt to generate a plan. The plan can include one or more actions that the copilot can take to complete the task. After the plan is generated

the copilot executes the actions and returns a response to the user." This document explicitly details the interpret

plan

and execute cycle.

2. Salesforce Developers Documentation

Einstein Copilot Actions: "Einstein Copilot uses a large language model (LLM) to interpret a user’s request and generate a plan of which actions to execute to fulfill the request. The LLM selects the best action or sequence of actions based on the action names and descriptions you provide." This source confirms that the LLM's core function is to generate a plan by selecting appropriate actions.

To route SMS text messages via Omni-Channel, a Service Channel must be configured for the MessagingSession object. Salesforce automatically creates a "Messaging" Service Channel when you set up a messaging channel (like SMS). This channel links incoming text messages to the Omni-Channel routing engine. An Omni-Channel flow then uses the Route Work action to direct these MessagingSession records to the appropriate agent, but the "Messaging" Service Channel is the foundational component that defines this type of work item for routing.

B. Route Work Action: This is an action within an Omni-Channel Flow that executes the routing logic. It is not the Service Channel itself, which defines the source object (e.g., MessagingSession) for the work.
C. Live Agent: This Service Channel is specifically for routing real-time web chats (represented by the LiveChatTranscript object), not for asynchronous SMS text messages.

1. Salesforce Help, "Create Service Channels": "A service channel lets you turn nearly any Salesforce object...into a work record. Omni-Channel then plucks these work records from their queues...and routes them to your agents... For channels like chat, messaging, and voice, service channels are created for you." This document confirms that "Messaging" is a distinct Service Channel type created automatically for channels like SMS.

2. Salesforce Help, "Route Messaging Sessions with Omni-Channel": "When you complete Messaging setup, a service channel and a basic routing configuration are created for you... The service channel controls the settings for the Messaging Session object." This source directly links the setup of SMS with the creation of a "Messaging" Service Channel for the MessagingSession object.

3. Salesforce Help, "Route Work with Omni-Channel Flows": This document details the Route Work action, describing it as the flow element that "routes a work item to a queue, agent, or skill." This distinguishes the action of routing from the definition of the work item, which is the role of the Service Channel.