<< Click to Display Table of Contents >> Navigation:  Low-code Process Automation > Studio Cloud - Authoring environment > Bizagi Studio > AI Hub > AI Agents

Overview

AI Agents bring the power of Generative Artificial Intelligence (Gen AI) directly to your business. These specialized assistants, configured similarly to Generative Pre-trained Transformers (GPT), leverage Azure Open AI to deliver tailored solutions based on natural language instructions, driving innovation and enhancing competitiveness.

 

AI Agents can target specific business groups to improve customer satisfaction, productivity, and employee engagement. They are useful for a variety of tasks, including but not limited to the following examples:

•Creation: Generate, categorize, and summarize content such as emails, documents, and articles.

•Summarization: Outline or condense lengthy meeting notes, transcripts, or documents.

•Obfuscation: Detect and mask personal information to ensure privacy.

•Categorization: Classify incoming requests, tickets, and cases.

•Translation: Convert data or text into multiple languages.

•Prioritization: Apply priority levels to requests, tickets, and cases.

•Analysis: Identify sentiment, tone, and logical groupings to gain data insights.

 

Before you start

Before using Bizagi's AI Agents, it's important to understand how they work and how to set them up effectively:

•Configuration Requirements: AI Agents need to be configured with input parameters and mapped outputs to your Data Model. Each output must be assigned a name and a data type. Supported data types include: String, Number, Date, Boolean, and Collection.

•AI Actions: AI Actions allow you to incorporate powerful, pre-configured connectors into your AI Agents without additional mapping. Simply describe your objective in natural language, and the system handles the rest.

•Enterprise Knowledge: This is a centralized framework for storing and managing business knowledge. It integrates with AI Agents to support better decision-making and improve the accuracy of AI-driven functionalities.Before using Bizagi's AI Agents, it is essential to understand their functionality:

•Categorical Structure: AI Agents are grouped in a node structure categorized by business units (e.g., Finance, Marketing, Sales) or projects. This structure supports only one level of categorization.

 

Configure an AI Agent

To use AI Agents in Bizagi Studio, follow these configuration steps:

1.Expert view configuration: Manage AI Agents and their Categories in the Expert view under the AI Hub option. You can create, edit, or remove both Categories and AI Agents. When creating an AI Agent, you have the option to use Templates or build it from scratch.

2.Activity Action configuration: In the Process Wizard's Business Rules step, configure an AI Agent as an Activity Action within an AI Task.

3.Form button configuration: In the Process Wizard's Define Forms step, configure a button in a form to execute an AI Agent.

4.Configuration options:

oTriggering options: You can trigger an AI Agent in two ways:

▪As an Activity Action within an AI Task (a task type exclusively for AI Agents).

▪From a button in a Form.

 

Using AI Agents

•Execute AI Agents from Activity Actions: Use the Activity Actions wizard to configure the AI Agent’s inputs, outputs, and error handling. Specify whether it runs at the task's start (On Enter) or finish (On Exit).

•Execute AI Agents from Form Actions: Invoke the ability of AI Agents to solve complex tasks by using Form Actions buttons that you configure from Actions and Validations.

•AI Agents from Entity Forms:  Leverage AI Agents directly within entity-based forms using the Execute AI Agents from Entity Forms feature.

•AI Agents in Pipelines: AI Agents can be integrated into pipelines, enabling automated processing of data at different stages. This enhances decision-making and streamlines workflows.

 

Last Updated 5/20/2025 3:00:48 PM