Customize Conversation Recommendations for Agentforce Employee Agents
Customize conversation recommendations to suggest prompts employees can use with your
Agentforce Employee agent, so they can select a prompt to start a conversation instead of typing
from scratch. Conversation recommendations include welcome recommendations that appear on the
welcome screen when the Agentforce panel opens, and in-conversation recommendations that the
agent suggests during a conversation.
Before you begin, set up Data Cloud. Data Cloud is required for
in-conversation recommendations only. Without it, in-conversation recommendations don't
appear, even when in-conversation recommendations are turned on. Welcome recommendations
don't require Data Cloud.
Customize Conversation Recommendations in Agentforce Builder Script View
To make changes to your agent, it must be in a draft state. If you want to make changes
to a committed agent version, create a draft.
From the App Launcher, search for Agent, and then select
Agentforce Studio.
On the Agents tab, click the name of the Agentforce Employee
agent that you want to customize.
In the Explorer panel, open Settings and click
System.
Switch to Script view. Under system, add a
recommended_prompts section by pasting this block into the
script.
recommended_prompts:
welcome_screen: True
in_conversation: True
starter_prompts:
- "Summarize Acme Account"
- "Get Details for Global Media"
- "Draft an email to Jon Amos"
Modify the pasted block for your agent.
To show welcome starter prompts on the welcome screen, keep
welcome_screen set to True. To hide them, set it to
False.
To show dynamic prompts that your agent suggests during a conversation, keep
in_conversation set to True. To hide them, set it
to False.
Replace the sample prompts under starter_prompts with your own.
Put each welcome starter prompt on its own line, start each line with a hyphen (-), and
enclose the prompt text in double quotation marks (").
Add at least three and up to twenty welcome starter prompts, with each prompt
containing 5–50 characters. Up to three prompts appear on the welcome screen at a time.
The order of the prompts in the list determines priority, but the prompts shown depend
on the page context.
Save your changes.
After you activate your agent, the configured prompts appear on the welcome screen when
the Agentforce panel opens, and the agent suggests dynamic prompts during a conversation in
response to the user's messages.
Customize Welcome Recommendations in the Legacy Builder
From Setup, in the Quick Find box, search for Agent, and select
Agentforce Agents.
Click the name of the Agentforce Employee agent you want to customize welcome
recommendations for.
From the Agent Details page, select the System Messages
tab.
Select Create Your Own and add your welcome
recommendations.
Add at least three and up to twenty welcome starter recommendations, with each prompt
containing 5–50 characters. Up to three prompts appear on the welcome screen at a time.
The order of the prompts in the list determines priority, but the prompts shown depend on
the page context.
Save your changes.
Did this article solve your issue?
Let us know so we can improve!
Loading
Salesforce Help | Article
Cookie Consent Manager
Cookie Consent Manager
General Information
Required Cookies
Functional Cookies
Advertising Cookies
General Information
We use three kinds of cookies on our websites: required, functional, and advertising. You can choose whether functional and advertising cookies apply. Click on the different cookie categories to find out more about each category and to change the default settings.
Privacy Statement
Required Cookies
Always Active
Required cookies are necessary for basic website functionality. Some examples include: session cookies needed to transmit the website, authentication cookies, and security cookies.
Functional Cookies
Functional cookies enhance functions, performance, and services on the website. Some examples include: cookies used to analyze site traffic, cookies used for market research, and cookies used to display advertising that is not directed to a particular individual.
Advertising Cookies
Advertising cookies track activity across websites in order to understand a viewer’s interests, and direct them specific marketing. Some examples include: cookies used for remarketing, or interest-based advertising.