Prompt Engineering
● Q
uestion 1:Universal Containers built a Field Generation prompt template that worked
for many records, but users are reporting random failures with token limit errors. What is
the cause of the random nature of this error?
○ A. The template type needs to be switched to Flex to accommodate the variable
amount of tokens generated by the prompt grounding.
○ B. The number of tokens generated by the dynamic nature of the prompt template
will vary by record.
○ C. The number of tokens that can be processed by the LLM varies with total user
demand.
● Q
uestion 5:Universal Containers (UC) is experimenting with using public Generative AI
models and is familiar with the language required to get the information it needs.
However, it can be time consuming for both UC's sales and service reps to type in the
prompt to get the information they need, and ensure prompt consistency. Which
Salesforce feature should the company use to address these concerns?
○ A. Einstein Prompt Builder and Prompt Templates
○ B. Einstein Recommendation Builder
○ C. Agent Builder and Action: Query Records
● Q
uestion 8:Universal Containers plans to enhance its sales team's productivity using AI.
Which specific requirement necessitates the use of Prompt Builder?
○ A. Creating a draft newsletter for an upcoming tradeshow
○ B. Predicting the likelihood of customers churning or discontinuing their relationship
with the company
○ C. Creating an estimated Customer Lifetime Value (CLV) with historical purchase
data
● Q
uestion 10:Universal Containers (UC) recently rolled out Einstein Generative AI
capabilities and has created a custom prompt to summarize case records. Users have
reported that the case summaries generated are not returning the appropriate information.
What is a possible explanation for the poor prompt performance?
○ A. The data being used for grounding is incorrect or incomplete.
○ B. The Einstein Trust Layer is incorrectly configured.
○ C. The prompt template version is incompatible with the chosen LLM.
● Q
uestion 13:Universal Containers wants to implement a solution in Salesforce with a
custom UX that allows users to enter a sales order number. Subsequently, the system will
invoke a custom prompt template to create and display a summary of the sales order
header and sales order details. Which solution should an Agentforce Specialist implement
to meet this requirement?
○ A. Create a template-triggered prompt flow and invoke the prompt template using
the standard "Prompt Template" flow action.
○ B. Create a screen flow to collect sales order number and invoke the prompt
template using the standard "Prompt Template" flow action.
C. Create an autolaunched flow and invoke the prompt template using the standard
○
"Prompt Template" flow action.
● Q
uestion 14:Universal Containers has an active standard email prompt template that
does not fully deliver on the business requirements. Which steps should an Agentforce
Specialist take to use the content of the standard prompt email template in question and
customize it to fully meet the business requirements?
○ A. Save as New Template and edit as needed.
○ B. Clone the existing template and modify as needed.
○ C. Save as New Version and edit as needed.
● Q
uestion 24:What considerations should an Agentforce Specialist be aware of when
using Record Snapshots grounding in a prompt template?
○ A. Empty data, such as fields without values or sections without limits, is filtered
out.
○ B. Activities such as tasks and events are excluded.
○ C. Email addresses associated with the object are excluded.
● Q
uestion 28:Universal Containers has grounded a prompt template with a related list.
During user acceptance testing (UAT), users are not getting the correct responses. What
is causing this issue?
○ A. The related list is not on the parent object's page layout.
○ B. The related list is Read Only.
○ C. The related list prompt template option is not enabled.
● Q
uestion 30:Universal Containers tests out a new Einstein Generative AI feature for its
sales team to create personalized and contextualized emails for its customers.
Sometimes, users find that the draft email contains placeholders for attributes that could
have been derived from the recipient's contact record. What is the most likely explanation
for why the draft email shows these placeholders?
○ A. The user's locale language is not supported by Prompt Builder.
○ B. The user does not have permission to access the fields.
○ C. The user does not have Einstein Sales Emails permission assigned.
● Q
uestion 31:An Agentforce Specialist needs to create a prompt template to fill a custom
field named Latest Opportunities Summary on the Account object with information from
the three most recently opened opportunities. How should the Agentforce Specialist
gather the necessary data for the prompt template?
○ A. Select the Account Opportunity object as a resource when creating the prompt
template.
○ B. Create a flow to retrieve the opportunity information.
○ C. Select the latest Opportunities related list as a merge field.
● Q
uestion 34:Universal Containers wants to leverage the Record Snapshots grounding
feature in a prompt template. What preparations are required?
○ A. Create a field set for all the fields to be grounded
○ B. Configure page layout of the master record type
○ C. Enable and configure dynamic form for the object
● Q
uestion 35:Universal Containers (UC) wants to make a sales proposal and directly use
data from multiple unrelated objects (standard and custom) in a prompt template. How
should UC accomplish this?
○ A. Create a prompt template passing in a special custom object that connects the
records temporarily.
○ B. Create a prompt template-triggered flow to access the data from standard and
custom objects.
○ C. Create a Flex template to add resources with standard and custom objects as
inputs.
● Q
uestion 36:When configuring a prompt template, an Agentforce Specialist previews the
results of the prompt template they've written. They see two distinct text outputs:
Resolution and Response. Which information does the Resolution text provide?
○ A. It shows which sensitive data is masked before it is sent to the LLM.
○ B. It shows the response from the LLM based on the sample record.
○ C. It shows the full text that is sent to the Trust Layer.
● Q
uestion 39:Universal Containers (UC) plans to implement prompt templates that utilize
the standard foundation models. What should UC consider when building prompt
templates in Prompt Builder?
○ A. Include multiple-choice questions within the prompt to test the LLM's
understanding of the context.
○ B. Train LLM with data using different writing styles including word choice,
intensifiers, emojis, and punctuation.
○ C. Ask it to role-play as a character in the prompt template to provide more context
to the LLM.
● Q
uestion 45: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?
○ A. Agent Builder
○ B. Model Builder
○ C. Prompt Builder
● Q
uestion 53:Universal Containers needs its sales reps to be able to only execute prompt
templates. What should the company use to achieve this requirement?
○ A. Prompt Execute Template permission set
○ B. Prompt Template User permission set
○ C. Prompt Template Manager permission set
● Q
uestion 56:Universal Containers (UC) is creating a new custom prompt template to
populate a field with generated output. UC enabled the Einstein Trust Layer to ensure AI
Audit data is captured and monitored for adoption and possible enhancements. Which
prompt template type should UC use and which consideration should UC review?
○ A. Field Generation, and that Dynamic Forms is enabled
B
○ . Flex, and that Dynamic Fields is enabled
○ C. Field Generation, and that Dynamic Fields is enabled
● Q
uestion 59:Universal Containers (UC) wants to enable its sales team to use AI to
suggest recommended products from its catalog. Which type of prompt template should
UC use?
○ A. Record summary prompt template
○ B. Email generation prompt template
○ C. Flex prompt template
● Q
uestion 60:A data scientist needs to view and manage models in Einstein Studio. The
data scientist also needs to create prompt templates in Prompt Builder. Which permission
sets should an Agentforce Specialist assign to the data scientist?
○ A. Prompt Template Manager and Prompt Template User
○ B. Data Cloud Admin and Prompt Template Manager
○ C. Prompt Template User and Data Cloud Admin
Agentforce Concepts
● Q
uestion 2:What is the role of the large language model (LLM) in understanding intent
and executing an Agent Action?
○ A. Identify the best matching topic and actions and correct order of execution
○ B. Find similar requested topics and provide the actions that need to be executed
○ C. Determine a user's topic access and sort actions by priority to be executed
● Q
uestion 4:An Agentforce Specialist is creating a custom action in Agentforce. Which
option is available for the Agentforce Specialist to choose for the custom Agent action?
○ A. Flows
○ B. SOQL
○ C. Apex trigger
● Q
uestion 7:Universal Containers recently added a custom flow for processing returns
and created a new Agent Action. Which action should the company take to ensure the
Agentforce Service Agent can run this new flow as part of the new Agent Action?
○ A. Assign the Run Flows permission to the Agentforce Agent user.
○ B. Assign the Manage Users permission to the Agentforce Agent user.
○ C. Recreate the flow using the Agentforce Agent user.
● Q
uestion 18:A customer service representative is looking at a custom object that stores
travel information. They recently received a weather alert and now need to cancel flights
for the customers that are related with this itinerary. The representative needs to review
the Knowledge articles about canceling and rebooking the customer flights. Which
Agentforce capability helps the representative accomplish this?
○ A. Execute tasks based on available actions, answering questions using information
from accessible Knowledge articles.
○ B. Generate a Knowledge article based off the prompts that the agent enters to
reate steps to cancel flights.
c
C. Invoke a flow which makes a call to external data to create a Knowledge article.
○
● Q
uestion 22:Universal Containers (UC) wants to ensure the effectiveness, reliability, and
trust of its agents prior to deploying them in production. UC would like to efficiently test a
large and repeatable number of utterances. What should the Agentforce Specialist
recommend?
○ A. Deploy the agent in a Q/A sandbox environment and review the Utterance
Analysis reports to review effectiveness.
○ B. Create a CSV file with UC's test cases in Agentforce Testing Center using the
testing template.
○ C. Leverage the Agent Large Language Model (LLM) UI and test UC's agents with
different utterances prior to activating the agent.
● Q
uestion 23:An Agentforce Specialist is tasked with analyzing Agent interactions looking
into user inputs, requests, and queries to identify patterns and trends. What functionality
allows the Agentforce Specialist to achieve this?
○ A. Agent Event Logs dashboard
○ B. AI Audit and Feedback Data dashboard
○ C. User Utterances dashboard
● Q
uestion 27:Universal Containers' 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?
○ A. The Apex classes may bypass the 75% code coverage requirement as long as
they are only used by the agent.
○ B. The Apex classes must have at least 75% code coverage from unit tests, and all
dependencies must be in the deployment package.
○ C. 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.
● Q
uestion 32:Universal Containers recently launched a pilot program to integrate
conversational AI into its CRM business operations with Agentforce Agents. How should
the Agentforce Specialist monitor Agents' usability and the assignment of actions?
○ A. Run a report on the Platform Debug Logs.
○ B. Query the Agent log data using the metadata API.
○ C. Run Agent Analytics.
● Q
uestion 43:How does an Agent respond when it can't understand the request or find
any requested information?
○ A. With a generated error message
○ B. With a general message asking the user to rephrase the request
○ C. With a preconfigured message, based on the action type
● Question 48:What is true of Agentforce Testing Center?
○ A. Running tests risks modifying CRM data in a production environment.
○ B. Running tests does not consume Einstein Requests.
○ C. Agentforce Testing Center can only be used in a production environment.
● Q
uestion 49:What should Universal Containers consider when deploying an Agentforce
Service Agent with multiple topics and Agent Actions to production?
○ A. Deploy agent components without a test run in staging, relying on production
data for reliable results. Sandbox configuration alone ensures seamless production
deployment.
○ B. Ensure all dependencies are included, Apex classes meet 75% test coverage,
and configuration settings are aligned with production. Plan for version
management and post-deployment activation.
○ C. Deploy flows or Apex after agents, topics, and Agent Actions to avoid
deployment failures and potential production agent issues requiring complete
redeployment.
● Q
uestion 52:What is the importance of Action Instructions when creating a custom
Agent action?
○ A. Action Instructions define the expected user experience of an action.
○ B. Action Instructions tell the user how to call this action in a conversation.
○ C. Action Instructions tell the large language model (LLM) which action to use.
● Q
uestion 57:An Agentforce Specialist wants to troubleshoot their Agent's performance.
Where should the Agentforce Specialist go to access all user interactions with the Agent,
including Agent errors, incorrectly triggered actions, and incomplete plans?
○ A. Event Logs
○ B. Agent Settings
○ C. Plan Canvas
● Q
uestion 58:Universal Containers implements Custom Agent Actions to enhance its
customer service operations. The development team needs to understand the core
components of a Custom Agent Action to ensure proper configuration and functionality.
What should the development team review in the Custom Agent Action configuration to
identify one of the core components of a Custom Agent Action?
○ A. Output Types
○ B. Instructions
○ C. Action Triggers
Agentforce and Data Cloud
● Q
uestion 3:Universal Containers (UC) implements a custom retriever to improve the
accuracy of AI-generated responses. UC notices that the retriever is returning too many
irrelevant results, making the responses less useful. What should UC do to ensure only
relevant data is retrieved?
○ A. Define filters to narrow the search results based on specific conditions.
○ B. Change the search index to a different data model object (DMO).
○ C. Increase the maximum number of results returned to capture a broader dataset.
● Question 6:Universal Containers (UC) uses a file upload-based data library and custom
rompt to support AI-driven training content. However, users report that the AI frequently
p
returns outdated documents. Which corrective action should UC implement to improve
content relevancy?
○ A. Configure a custom retriever that includes a filter condition limiting retrieval to
documents updated within a defined recent period, ensuring that only current
content is used for AI responses.
○ B. Switch the data library source from file uploads to a Knowledge-based data
library, because Salesforce Knowledge bases automatically manage document
recency, ensuring current documents are returned.
○ C. Continue using the default retriever without filters, because periodic re-uploads
will eventually phase out outdated documents without further configuration or the
need for custom retrievers.
● Q
uestion 11:Universal Containers (UC) has configured Agentforce Data Library using
Knowledge articles. When testing in Agent Builder and the Experience Cloud site, the
agent is not responding with grounded Knowledge article information. However, when
tested in Prompt Builder, the response returns correctly. What should UC do to
troubleshoot the issue?
○ A. Create a new permission set that assigns "Manage Knowledge" and assign it to
the Agentforce Service Agent User.
○ B. Ensure the assigned User permission set includes access to the prompt template
used to access the Knowledge articles.
○ C. Ensure the Data Cloud User permission set has been assigned to the Agentforce
Service Agent User.
● Q
uestion 12:Universal Containers (UC) wants to build an Agentforce Service Agent that
provides the latest, active, and relevant policy and compliance information to customers.
The agent must: • Semantically search HR policies, compliance guidelines, and company
procedures. • Ensure responses are grounded on published Knowledge. • Allow
Knowledge updates to be reflected immediately without manual reconfiguration. What
should UC do to ensure the agent retrieves the right information?
○ A. Enable the agent to search all internal records and past customer inquiries.
○ B. Set up an Agentforce Data Library to store and index policy documents for AI
retrieval.
○ C. Manually add policy responses into the AI model to prevent hallucinations.
● Q
uestion 15:For an Agentforce Data Library that contains uploaded files, what occurs
once it is created and configured?
○ A. Indexes the uploaded files into Data Cloud
○ B. Indexes the uploaded files in Salesforce File Storage
○ C. Indexes the uploaded files in a location specified by the user
● Question 16:What is a valid use case for Data Cloud retrievers?
○ A. Returning relevant data from the vector database to augment a prompt
○ B. Modifying and updating data within the source systems connected to Data Cloud
○ C. Grounding data from external websites to augment a prompt with RAG
● Question 17:Universal Containers would like to route SMS text messages to a service
r ep from an Agentforce Service Agent. Which Service Channel should the company use
in the flow to ensure it's routed properly?
○ A. Messaging
○ B. Route Work Action
○ C. Live Agent
● Q
uestion 19:What is automatically created when a custom search index is created in
Data Cloud?
○ A. A dynamic retriever to allow runtime selection of retriever parameters without
manual configuration
○ B. A retriever that shares the name of the custom search index
○ C. A predefined Apex retriever class that can be edited by a developer to meet
specific needs
● Question 26:How does the AI retriever function within Data Cloud?
○ A. It monitors and aggregates data quality metrics across various data pipelines to
ensure only high-integrity data is used for strategic decision-making.
○ B. It performs contextual searches over an indexed repository to quickly fetch the
most relevant documents, enabling grounding AI responses with trustworthy,
verifiable information.
○ C. It automatically extracts and reformats raw data from diverse sources into
standardized datasets for use in historical trend analysis and forecasting.
.
● Question 29:Universal Containers (UC) is rolling out an AI-powered support assistant to
help customer service agents quickly retrieve relevant troubleshooting steps and policy
guidelines. The assistant relies on a search index in Data Cloud that contains product
manuals, policy documents, and past case resolutions. During testing, UC notices that
agents are receiving too many irrelevant results from older product versions that no longer
apply. How should UC address this issue?
○ A. Modify the search index to only store documents from the last year and remove
older records.
○ B. Create a custom retriever in Einstein Studio, and apply filters for publication date
and product line.
○ C. Use the default retriever, as it already searches the entire search index and
provides broad coverage.
● Q
uestion 37:When creating a custom retriever in Einstein Studio, which step is
considered essential?
○ A. Define the output configuration by specifying the maximum number of results to
return, and map the output fields that will ground the prompt.
○ B. Configure the search index, choose vector or hybrid search, choose the fields for
filtering, the data space and model, then define the ranking method.
○ C. Select the search index, specify the associated data model object (DMO) and
data space, and optionally define filters to narrow search results.
● Q
uestion 41:Universal Containers (UC) wants to limit an agent's access to Knowledge
articles, while deploying Answer Questions with Knowledge action. How should UC
achieve this?
○ A . Define scope instructions to the agent specifying a list of allowed article titles or
IDs.
○ B. Assign Data Categories to Knowledge articles, and define Data Category filters
in the Agentforce Data Library.
○ C. Update the Data Library Retriever to filter on a custom field on the Knowledge
article.
● Q
uestion 42:In a Knowledge-based data library configuration, what is the primary
difference between the identifying fields and the content fields?
○ A. Identifying fields categorize articles for indexing purposes, while content fields
provide a brief summary for display.
○ B. Identifying fields help locate the correct Knowledge article, while content fields
enrich AI responses with detailed information.
○ C. Identifying fields highlight key terms for relevance scoring, while content fields
store the full text of the article for retrieval.
● Q
uestion 54:Which scenario best demonstrates when an Agentforce Data Library is
most useful for improving an AI agent's response accuracy?
○ A. When data is being retrieved from Snowflake using zero-copy for vectorization
and retrieval.
○ B. When the AI agent needs to combine data from disparate sources based on
mutually common data, such as Customer Id and Product Id for grounding.
○ C. When the AI agent must provide answers based on a curated set of policy
documents that are stored, regularly updated, and indexed in the data library.
● Q
uestion 55:Which element in the Omni-Channel flow should be used to connect the
flow with the agent?
○ A. Assignment
○ B. Route Work Action
○ C. Decision
Agentforce and Service Cloud
● Q
uestion 20:Universal Containers (UC) wants to leverage Generative AI Salesforce
functionalityto reduce Service Agent handling time by providing recommended replies
based on the existing Knowledge articles. On which AI capability should UC train the
service agents?
○ - A. Service Replies
○ - B. Knowledge Replies
○ - C. Case Replies
● Q
uestion 38:Universal Containers wants to reduceoverall customer support handling
time by minimizing the time spent typing routine answers for common questions in-chat,
and reducing the post-chat analysis by suggesting values for case fields. Which
combination of Agentforce for Service features enables this effort?
○ A. Einstein Service Replies and Work Summaries
○ B. Einstein Reply Recommendations and Case Classification
○ C. Einstein Reply Recommendations and Case Summaries
● Q
uestion 47:Universal Containers deploys a new Agentforce Service Agent into the
company's website but is getting feedback that the Agentforce Service Agent is not
providing answers to customer questions that are found in the company's Salesforce
Knowledge articles. What is the likely issue?
○ A. The Agentforce Service Agent user is not assigned the correct Agent Type
License.
○ B. The Agentforce Service Agent user needs to be created under the standard
Agent Knowledge profile.
○ C. The Agentforce Service Agent user was not given the Allow View Knowledge
permission set.
● Q
uestion 50:Universal Containers' service team wants to customize the standard case
summary response from Agentforce. What should the Agentforce Specialist do to achieve
this?
○ A. Create a custom Record Summary prompt template for the Case object.
○ B. Customize the standard Record Summary template for the Case object.
○ C. Summarize the Case with a standard Agent action.
● Q
uestion 51:Universal Containers has implemented an agent that answers questions
based on Knowledge articles. Which topic and Agent Action will be shown in the Agent
Builder?
○ A. General FAQ topic and Answers Questions with Knowledge Action
○ B. General CRM topic and Answers Questions with LLM Action
○ C. General Q&A topic and Knowledge Article Answers action
Agentforce and Sales Cloud
● Q
uestion 9:Amid their busy schedules, sales reps at Universal Containers dedicate time
to follow up with prospects and existing clients via email regarding renewals or new deals.
They spend many hours throughout the week reviewing past communications and details
about their customers before performing their outreach. Which standard Agent action
helps sales reps draft personalized emails to prospects by generating text based on
previous successful communications?
○ A. Agent Action: Summarize Record
○ B. Agent Action: Find Similar Opportunities
○ C. Agent Action: Draft or Revise Sales Email
● Q
uestion 21:Universal Containers is using Agentforce for Sales to find similar
opportunities to help close deals faster. The team wants to understand the criteria used by
the Agent to match opportunities. What is one criteria that Agentforce for Sales uses to
match similar opportunities?
○ A. Matched opportunities are limited to the same account.
○ B. Matched opportunities were created in the last 12 months.
○ C. Matched opportunities have a status of Closed Won from last 12 months.
● Q
uestion 25:Universal Containers (UC) currently tracks Leads with a custom object. UC
is preparing to implement the Sales Development Representative (SDR) Agent. Which
consideration should UC keep in mind?
○ A. Agentforce SDR only works with the standard Lead object.
○ B. Agentforce SDR only support custom objects associated with Accounts.
○ C. Agentforce SDR only works on Opportunites.
● Q
uestion 33:Universal Containers (UC) would like to implement Sales Development
Representative (SDR) Agent. Which channel consideration should UC be aware of while
implementing it?
○ A. SDR Agent must also be deployed in the company website.
○ B. SDR Agent only works in Email channel.
○ C. SDR Agent must be deployed in Messaging channel.
● Q
uestion 40:Universal Containers (UC) wants to implement an AI-powered customer
service agent that can: • Retrieve proprietary policy documents that are stored as PDFs. •
Ensure responses are grounded in approved company data, not generic LLM knowledge.
What should UC do first?
○ A. Expand the AI agent's scope to search all Salesforce records.
○ B. Set up an Agentforce Data Library for AI retrieval of policy documents.
○ C. Add the files to the content, and then select the data library option.
● Q
uestion 44:Universal Containers (UC) wants to enable its sales team to get insights
into product and competitor names mentioned during calls. How should UC meet this
requirement?
○ A. Enable Einstein Conversation Insights, assign permission sets, define recording
managers, and customize insights with up to 50 competitor names.
○ B. Enable Einstein Conversation Insights, connect a recording provider, assign
permission sets, and customize insights with up to 25 products.
○ C. Enable Einstein Conversation Insights, enable sales recording, assign
permission sets, and customize insights with up to 50 products.
● Q
uestion 46:Universal Containers wants to utilize Agentforce for Sales to help sales
reps reach their sales quotas by providing AI-generated plans containing guidance and
steps for closing deals. Which feature meets this requirement?
○ A. Find Similar Deals
○ B. Create Account Plan
○ C. Create Close Plan
