Skip to main content
Topics#Mvp-agentforce-prompt-builder-team

#Mvp-agentforce-prompt-builder-team8 discussing

Ask a question...
David Königbauer (Your Company !) asked in #Trailhead Challenges
Yesterday, 10:51 AM
Prompt Builder - Template - Task - Fail

Hi Community. 

 

I hope you are doing well.   

 

I receive an error when checking the task in this trail:

https://trailhead.salesforce.com/de/content/learn/projects/quick-start-prompt-builder/get-started-with-prompt-builder

:  

 

Error:  

Step in Einstein AI not yet complete 

We cannot find an active prompt template for field generation named 'Quick Summary'. Ensure that the API name of the prompt template is 'Quick_Summary' 

 

But there is an active prompt template "Quick_Summary" which I created before.  

 

Does anybody have a clue what causes the issue? 

 

Many Thanks & Regards 

David  

 

#Mvp-agentforce-prompt-builder-team

  

 

#Trailhead Challenges

1 answer
  1. David Königbauer (Your Company !)
    Yesterday, 1:45 PM

    Hi Tushar, thank you for your quick response. Unfortunately it didn´t help. The screenshot shows that there is nothing wrong. 

     

    Have a good weekend,  

    David  

    Hi Tushar, thank you for your quick response. Unfortunately it didn´t help. The screenshot shows that there is nothing wrong.

     

     

Write an answer...
0/9000
NARASIMHADEVARA VAISHNAVI SRIYAA (VNRVJIET) asked in #Agentforce
Oct 20, 1:50 PM
My web agent thought my fix was its own work.

The loop that hit me hard:Web Agent types wrong value in field. 

I manually correct it. 

Web Agent continues: "Great, I typed it correctly" → moves to next step. 

Everything after that fails because it thinks it made my edit. It had no idea I touched the browser at all. Built ActionTrace to fix this exact blind spot: 

[AI] typed "ABC" → wrong 

[HUMAN] typed "XYZ" → this is the real value now 

Now the Web Agent knows that an error didn't happen magically, human made the change - without human needing to explicitly say it. 

🔗 GitHub:

https://github.com/nvaishnavisriyaa/ActionTrace

 

🎥 45-sec demo:

https://youtu.be/WKw3Eqf8n_E

 

Question: When human edits a field mid-flow in Service Cloud, does Agentforce get live context that "human made this change"? Or does it assume it made the change itself? Anyone seen this? 

 

#Agentforce  #AI  #AgenticAI  #AgentforceBuilder  #Mvp-agentforce-prompt-builder-team

1 answer
  1. dhanush gunda (singingbirdapps.com)
    Yesterday, 11:19 AM

    Nice find. From what I've seen, you shouldn't assume Agentforce tracks who changed a field between steps. Agents reason over what their actions return and what's in the conversation, so if nothing re-reads the record, the agent's view is whatever it last saw or did. 

    Patterns I'd use in Service Cloud: 

     

    1. Re-read before acting. Design actions (Flow or Apex) to query the current record state at the start of each step rather than relying on what the agent did earlier. 

     

     

    2. Make "who changed it" data. Turn on Field History Tracking for the key fields, and/or return LastModifiedById in your action, so the agent can see that a human edited the field. Then tell it in the instructions to treat a human edit as the source of truth. 

     

    3. Add a verification step. After the agent writes a value, have an action confirm the saved value matches what it intended before it moves on. 

     

    Your ActionTrace idea basically does #2 at the browser level. In the platform, the reliable version is to expose the change history through an action the agent can call.

Write an answer...
0/9000
Sourabh Dondekar (Jr. Developer at digiCloud Solutions Private Limited) asked in #Agentforce
Oct 14, 1:04 PM
Is Flex Consumption visible only in Production, and not in a Sandbox org?

In Production, Flex Consumption is visible under Digital Wallets. In my Sandbox, it does not appear. The Sandbox only shows the list view of “All Digital Wallets” and “Recently Viewed.”

Does Flex Consumption show only in Production? Or is it available in a Sandbox as well, but requires some configuration or setup to become visible?

Any guidance on what needs to be enabled or configured to see Flex Consumption under Digital Wallets in a Sandbox would be appreciated. 

 

#Agentforce  #AgentforceWorldTour  #AgentforceBuilder  #Mvp-agentforce-prompt-builder-team  #Prompt Builder

1 answer
  1. Venkat Yadav (SFDCINDIA)
    Oct 14, 2:05 PM

    Hi @Sourabh Dondekar Yes, this is expected. Flex Credit consumption is viewed through Digital Wallet in Production, even for Sandbox usage.

    A few key points:

    1. Digital Wallet consumption details are available in Production.
    2. Sandbox usage can be tracked from the Production Digital Wallet.
    3. Make sure you have the required permissions to view consumption details.

    Check Digital Wallet in your Production org. You don't need to enable any separate setting in Sandbox.

    Hope this helps!

Write an answer...
0/9000
Sourabh Dondekar (Jr. Developer at digiCloud Solutions Private Limited) asked in #Agentforce
Oct 12, 4:52 AM
Can we control Flex Credit consumption when invoking Prompt Templates from Record-Triggered Flows?
  • We have limited Flex Credits and want to make them available for all users in our Salesforce org.
  • We have created a Prompt Template and are invoking it from a Record-Triggered Flow.
  • We want to execute the prompt under a specific user context instead of allowing every user to execute it independently.
  • If we assign the Prompt Template User permission to a specific user and configure the flow to run in that user's context, will the prompt execute successfully?
  • Will this approach help us control or reduce Flex Credit consumption when multiple users trigger the flow?
  • Are there any licensing or permission requirements, such as the AI unmetered user license or Prompt Template User permission, that could prevent the prompt from executing?
  • Is there a recommended Salesforce-supported approach to allow all users to benefit from the prompt while managing limited Flex Credits?
  • Are there alternative solutions, such as a dedicated integration user, asynchronous Flow execution, or another automation design, to achieve this?

We would appreciate guidance on the recommended approach, including any limitations or best practices for managing Flex Credits when Prompt Templates are invoked through Record-Triggered Flows.

 

 

#Agentforce  #Prompt Builder  #Mvp-agentforce-prompt-builder-team

1 answer
  1. Venkat Yadav (SFDCINDIA)
    Oct 12, 11:02 AM

    Hi @Sourabh Dondekar Good question! Using a dedicated user for Prompt Template execution can help manage access, but it won't directly reduce Flex Credit consumption.

    A few suggestions:

    1. Flex Credits: Usage depends on billable AI executions, not the number of users.
    2. Running User: Record-Triggered Flows don't allow you to directly select a specific running user.
    3. Permissions: Make sure the execution context has the required Prompt Template permissions and access.
    4. Flow Optimization: Trigger the prompt only when necessary, using entry conditions and field-change checks.
    5. Asynchronous Flow: Helps manage processing but doesn't reduce credit consumption by itself.
    6. Monitoring: Use Digital Wallet to monitor usage and identify unnecessary executions.

    My suggestion: Focus on reducing unnecessary prompt calls rather than restricting users. You can also consider a centralized automation approach for better control.

    Hope this helps!

Write an answer...
0/9000
Sourabh Dondekar (Jr. Developer at digiCloud Solutions Private Limited) asked in #Agentforce
Sep 15, 7:11 AM
Does a user need the Prompt Template User permission to execute a Prompt Template from a Record-Triggered Flow ?

I have a Prompt Template that is invoked from a Record-Triggered Flow.

The user who triggers the record-triggered Flow does not have:

  • Prompt Template User permission 

I would like to understand the expected Salesforce behavior:

  1. Is the Prompt Template User permission required when a Prompt Template is invoked through a Record-Triggered Flow?
  2. If the triggering user does not have this permission, will the Prompt Template still execute?
  3. If the Prompt Template executes successfully, will it consume Flex Credits?
  4. Which user's permissions are evaluated when the Flow invokes the Prompt Template—the Flow-running user or the automation/system context?
  5. What is the purpose of the Prompt Template User permission if Prompt Templates can be invoked through Flow?

 

Looking for clarification based on the current Salesforce behavior/documentation for Record-Triggered Flow → Prompt Template

. 

 

#Agentforce  #Prompt Builder  #Mvp-agentforce-prompt-builder-team  #AgentforceTesting

8 answers
Write an answer...
0/9000