This topic describes how to create and manage an AI Agent in the Control Room. You can perform various actions such as running, moving, checking in, viewing, editing, copying, and deleting AI Agent, with certain permissions required for editing and deletion.

To create an AI Agent in the Control Room:

Procedure

  1. Log in to the Control Room and navigate to Automation > Create new or ‘+’ icon and choose AI Agent.
  2. Provide a name and description and click Create & edit to display a template outline. Let us create an example -- Reserve a Ride AI Agent.
  3. Define the agent's prompt: You will notice Generate prompt for your Agent (cloud-only feature) screen appears and ask you to input the prompt.

    The first step in creating an AI Agent is to give it a mission. You will do this by providing a clear and detailed initial screen prompt that describes the agent's purpose in natural language. This prompt is the foundation upon which the entire agent's logic is built.

    For example, to create the "Reserve a Ride" agent shown in the example, a user might have started with a concise but informative message like this:

    You work at an ride share company. Your job is to get the city name and should be able to book a ride for a user. That includes:
* getting the details including the city name.
* checking the weather first to decide if a car or a bike is a better option. 
* reserving the ride and confirm the details with the user.

    This initial input provides the core requirements, which the AI Agent Studio can then use to generate a more detailed, structured prompt. The AI Agent Studio is designed to be flexible; you do not need to provide every single detail upfront. However, the more context you provide, the better the AI Agent Studio can generate a starting template for you to refine.

    The system will then expand on this initial input to create a comprehensive prompt:

    Role: A transportation booking assistant specializing in dynamic ride reservations, leveraging real-time weather data and user preferences to ensure safe, efficient, and user-centric ride selection.

    Goal: Your goal is to dynamically reserve a ride for the user by selecting a car if it is raining or a bike if it is not raining, ensuring the reservation aligns with the user's preferences and providing a detailed confirmation of the booking, including ride type, pickup time, and location.

    AI Agent- Reserve Ride
  4. Choose a Model Connection:Navigate to Model tab and click Choose to choose from the available list of Model connections you have access to. For more information on Model Connection, see Model connections
    These Model connections are created by the Automation Admin and assigned to your user with a custom role.
    Note: When you use the gpt-4o model, select the November 2024 version or a later one. This instruction applies when you are selecting a model from a third-party source like Azure, as earlier versions are known to have issues. If you are using a model provided directly by Automation Anywhere, you will not have a choice of versions.
  5. Define agent variables: Navigate to Input & Output tab. Here, you need to tell the agent what information to look for as input variables and what to provide as output variables. These variables will be used by the LLM to understand the context and communicate with the tools. The key is providing rich and descriptive language. The more detailed your descriptions, the better the LLM will be at understanding its purpose and fulfilling its goal.

    • Input variable:

      • Name:details

      • Description: details of the reservation

    • Output variable:

      • Name:reservation_details

      • Description: final details of the reservation

  6. Define the End-user Display: This step allows you to configure how the AI Agent execution is presented to the user. The goal is to provide a clear and meaningful title for the task that the user sees in the Co-pilot interface. Navigate to the End-user display tab. In the Agent request title, provide a title for the task. This title is a mandatory field and is displayed when the AI Agent runs in the Co-pilot. For example, for the Reserve a Ride agent, the title could be Reserve Ride. This helps the user understand what the agent is doing at a glance.
  7. Add and configure tools:

    Under Tools tab, add the required automations for your agent to perform its task. In this example, you will be adding the WeatherChecker and ReserveRideBikeorCar automations as tools.

    Tool Name Purpose Input Variables Output Variables
    WeatherChecker Checks the current weather for a specified city. The LLM uses this tool to determine if it is raining. Name: city (which is part of the details input)

    Description: The city to check the current weather for. This is a mandatory input required to run the weather check.

    		 Name: weather_status

    Description: The weather status, such as 'Raining' or 'Sunny'. This output is critical for the agent's decision-making process.
    ReserveRideBikeorCar Books a ride (either a car or a bike) based on the specified ride type. The LLM calls this tool after it has made a decision based on the weather. Name: ridetype

    Description: The type of ride to book, which can be either 'Car' or 'Bike'. The LLM must determine this value based on the WeatherChecker tool's output.

    		Name: summary

    Description: The final, detailed confirmation summary of the reserved ride. This information is used to populate the agent's final output variable.

    For more information on configuring tools, see Configuring settings in Tools for an AI Agent.

  8. Define the Action Plan: The Action Plan section, is generated by the system based on your prompt and the tools you have added. It outlines the logical flow:

    • Retrieve weather data: Call the WeatherChecker tool using the city extracted from the input details.

    • Determine ride type: Based on the WeatherChecker's output, decide whether to book a car or a bike.

    • Reserve ride: Call the ReserveRideBikeorCar tool with the determined ridetype.

    • Confirm with the user: Use the output from the ReserveRideBikeorCar tool to provide a confirmation message to the user, which will become the reservation_details output variable.

  9. Test and refine: Once you have configured everything, click Run to test your agent to ensure it behaves as expected. You can simulate a conversation with the agent, providing a city, and seeing if it correctly checks the weather and books the appropriate ride type. Refine the variable descriptions and tool prompts as needed to improve the agent's performance and accuracy.

Next steps

Once you have created and configured your AI Agent, you can integrate it into your business processes and make it available to other users. The following diagram illustrates the typical workflow, from agent creation to execution in a public environment.

AI Agent Work flow

The process of moving an AI Agent from a development environment to a public-facing one involves several key steps and user roles.

Create and Test
A Professional Developer or Automation Admin creates and tests the AI Agent.
Configure in a Process
The Professional Developer then configures the AI Agent within a larger process.
Check In
The process is checked in to the repository, making it available for use.
Run in Public
The process can then be run in a public workspace, allowing it to be called through the Automation Co-Pilot.
Note:

A Business User with the required Automation Co-Pilot Business User license can execute the process and interact with the AI Agent. The diagram also shows the specific permissions required for each user role to perform these tasks. For more information about the roles and permissions, refer Roles and permissions.