Working with Action Apps and Agents

Terminology:

• Agents: Role-based, goal-oriented capabilities that autonomously make informed decisions and actions based on reason, context, and data. Best suited for ad-hoc processes requiring reasoning skills and high levels of adaptability. Agents dynamically use available business tools to retrieve information, determine necessary actions, and direct appropriate resources, including automation robots, to complete enterprise processes
• Action Apps: A type of apps designed in UiPath Apps, allowing you to create forms or UI interactions with specific options for tasks requiring human intervention (Action Center tasks), such as approvals. You can define the input and output properties of action apps by adding action properties.

Agents can encounter different problems. A tool call can fail unexpectedly or return data of an unexpected structure, they might need help discerning between multiple customers with the same name in a third-party app, or they might have low confidence in an action. There might also be organizational reasons that they aren’t trusted to make certain types of decisions or take certain actions.

In all of these cases, agents need to be able to escalate to a human and

1. surface all relevant details in an effective and straightforward way
2. allow the Human-In-The-Loop (HITL) to provide the relevant decision or information, whether this is via a simple Approval/Denial decision for a case or via providing several business details - emails, order numbers, or references to accounts in external platforms.

Agents might even need for one of their tools to be fixed before they can continue operation on a business case.

Tutorial - Using Action Apps for Agent Escalations

First you need to create and configure a new app.

Tutorial - Using Action Apps for Agent Escalations

First you need to create and configure a new app.

Note: We recommend you use Action Apps in Web Apps projects, rather than standalone Apps.

1. Go to Automation Cloud™ and select the Studio Web service.
2. From the Automations tab, select New project, and choose the App project type.
   
   
   
   
3. Select a preferred template, style, and the desired controls. For the purposes of this tutorial, use the Form D template.
   Read more about designing your app.
4. Select App name, then, under the General tab, select App name to open the expression editor.
5. Give the new app a name: "Tool Approval App" and save.
6. From the left navigation panel, select FormWrapper and remove the controls that you are not going to use in this example: FirstName, LastName, Email, Phone, Street, State, Zip, and Country.
7. Select the project explorer button on the top-left of the page, then select the + add button, and choose the Action option.
   
   
   
   
8. From the Toolbox button on the top-left of the page, navigate to the Display tab, and add a label to the app.
   
   
   
   

Now you need to turn this App into an Action App.

9. Select the project explorer button on the top-left of the page, then select the + add button, and choose the Action option. The Action properties dialog displays.
10. In the Action properties dialog, add webSearchResults in the Input Properties and make it Mandatory by checking the related box.
11. Create two outcomes: Approve and Deny.
    
    
    
    
12. Return to the form and name the label you previously created in the expression editor with "webSearchResults for approval is: "+ActionProperties.webSearchResults. Select Save.
    
    
    
    
13. Change the names of your Submit and Cancel buttons into Approve and Deny.
14. Select the Approve button and go to the Events tab on the right.
15. Add a "Clicked on" automation by selecting the Edit automation button.
16. Add a Submit Action activity.
    Read more about adding activities in Studio Web.
17. In the Outcome section, choose the Approve value.
18. Select the Deny button and go to the Events tab on the right.
19. Add a "Clicked on" automation by selecting the Edit automation button.
20. Add a Submit Action activity.
    Read more about adding activities in Studio Web.
21. In the Outcome section, choose the Deny value.

To make this Action App available in the organization, you need to publish and deploy it.

22. Select Publish.
23. Choose the name "EscalationTestApp", and select Publish again.
24. Go to Orchestrator and deploy the app. Make sure it runs on a serverless runtime.
25. After deploying the app, select the Edit the process details button.
26. Select Next.
27. In the Cloud Robot - Serverless section, choose the Medium option. This is the option we recommend for best performance with Web Apps.
28. Select Update.
29. Create a new Agent following this agent building procedure.
    1. In the System prompt input the following: Use the web search tool to do a search on the provided search item. You must then raise an escalation to confirm that the search result is satisfactory before returning your final output, a simple one-liner from the web search.
    2. In the User prompt field, input the following: Do a search on term: and select Arguments on the right of the prompt section.
       
       
       
       
    3. On the Arguments screen, in the Input schema tab, select Add Root Property.
    4. in the Name field input searchTerm, and in the Description, input the search term provided from the user. make this toot property Required.
       
       
       
       
    5. On the Arguments screen, in the Output schema tab, input the name output and the description natural language summary of the approved search results. Make this property Required.
    6. Select Confirm.
       
       
       
       
    7. 
       
    8. In the Tools section, choose a Web Search activity.
    9. In the UiPath GenAI Activities connection section, select a connection.
    10. Use a GoogleCustomSearch engine.
    11. In the Search description section, input The term to search the web for and, from the tune menu on the right, select Agent.
    12. In the Playground section on the right, in the searchTerm field, input news this week.
        
        
        
        
    13. Run the agent.
        The agent returns the required output.
        

        

        
        
    14. In the Escalations and Agent memory section, in the Description field, input Use this before using the Order Details tool.
    15. From the App templates section, select your app.
        
        
        
        
    16. Run the agent.
30. From Automation Cloud, go to Actions (previously Action Center) to see your pending tasks.
    1. Review the escalation task awaiting you and approve it if everything is in order.
       
       
       
       
31. Go back to the Agents workspace, where you should see the Agent executing the task and providing the expected output.

Escalations are how Agents? ask “what do I do now?”. Humans help resolve or complete escalations to get agents unstuck. The capability to raise escalations in a variety of different situations is a core factor that allows agents to address opportunities out of reach of traditional RPA.