Skip to main content
Feedback

Jumpstart your journey with Boomi Agentstudio

Dive into this guide to get started and learn the basics of Agentstudio .

What is Agentstudio?

Agentstudio is a full agent lifecycle management solution that enables organizations to design, govern, and orchestrate AI agents at scale with built-in governance for control, oversight, and compliance.

AI management framework

Agentstudio includes capabilities across the AI development and management lifecycle, comprised of three main components:

  • Agent Designer - Design and deploy AI agents quickly with intuitive low-code templates and tools in the Agent Designer. Seamlessly connect them to trusted data and ensure responsible development with built-in guardrails.
  • Agent Garden - Manage and interact with AI agents effortlessly using the Agent Garden’s intuitive interface. Organize and deploy created agents, and streamline the AI lifecycle.
  • Agent Control Tower - Maintain control over AI agents with the Agent Control Tower, the governance layer of Agentstudio. Proactively mitigate security and privacy risks while ensuring trust and compliance.

What are AI agents?

AI agents are autonomous or semi-autonomous software systems that achieve a defined goal by observing environments, following instructions, using reasoning, and taking action. To learn more about what AI agents are and how they can accelerate productivity and simplify your integration and automation initiatives, refer to About AI agents

Access

Agentstudio has several roles to help you control access to different actions and screens in Agentstudio.

  • Developer role - lets you create and deploy your own custom agents in the Agent Designer.
  • User role - lets you interact with deployed agents built by Agent Designer and any pre-installed Boomi AI agents available in the conversational user interface.
  • Administrator role - gives you access to all actions in Agentstudio, including the ability to delete agents.

Refer to Access and privileges to learn more.

Agent lifecycle

In Agentstudio, you can build, test, deploy, and manage AI agents built in Agent Designer. You can also manage agents from other sources, such as AWS and Salesforce.

diagram of four steps in agent management

Building AI Agents

You can create custom, goal-driven AI agents using Agent Designer to accelerate automation and productivity within your organization.

To start building agents you can:

Agent templates

If you are not sure what to build or how to build it, explore the templates gallery on the Agent Garden home page. There are two types of templates available:

  • Boomi agent templates - Install configurable and read-to-deploy agents built by Boomi. Refer to Building an agent from a template to learn more.
  • Marketplace agent templates - Install configurable agents built by Technology Partners and Boomi Community members. Refer to Installing Marketplace agents to learn more.

Agent examples

Agent examples walk you through building an agent step-by-step. They help you understand how to use Agent step, import MCP tools, and more with agents you build. Explore our agent examples collection to learn more.

Agent components

An AI Agent has several components that guide its behavior and functionality.

AI agent component diagram

  • Goal - The overarching objective guiding the agent's actions

  • Model - A Large Language Model (LLM) powers an agent's reasoning and generates output in natural language

  • Task - Specific actions undertaken to achieve the goal

  • Instructions - Prompts guiding the agent's behavior. Refer to Writing tasks and instructions for best practices and guidance.

  • Personality - The agent's voice and interaction style

  • Guardrails - Rules preventing inappropriate actions. Refer to Creating guardrails to learn more.

  • Tools - Specialized functions enhancing capabilities. Refer to Identifying and adding capabilities to agents to learn more.

MCP support

MCP (Model Context Protocol) is a universal interface that allows AI agents and large language models to connect to enterprise tools in a consistent way, making AI agent development much simpler. It works as a broker, exposing tools like APIs, databases, or file systems through MCP servers that define how agents can interact with them. By centralizing connections, MCP eliminates the need for one-off integrations, making it far easier to scale as the number of agents and tools grows.

With Agentstudio’s MCP support, you can instantly connect a wide range of tools without configuration, accelerating the development of intelligent automation within the Boomi Platform. Explore the following topics to help you get started:

Testing agents

You can build and test your agents using the Test Agent window in Agent Designer. As you interact with your agent, you can view the trace to see its reasoning and tool calls. It allows you to see why errors and latency occur so you can make adjustments to tasks, instructions, tools, and guardrails.

Refer to Testing and troubleshooting an agent to learn more about testing best practices.

Deploying agents

Once you have built and tested out your agent, you can deploy your agent for use in the Agent Garden. A deployed agent syncs to Agent Control Tower so you can manage it.

To deploy your agent, navigate to Agent Garden > Agents > select the Actions icon > select Deploy.

If you plan to use an agent in an Agent step for your integration, you must deploy the agent. After the agent syncs in Agent Control Tower, it will display as an option for Agent step.

You must disable your agent before you can edit it.

Managing and monitoring agents

Using the Agent Control Tower, you can manage and monitor your Boomi Agents alongside third-party providers like Amazon Bedrock and Salesforce. This management dashboard allows you to keep track and manage your created agents across different creators.

Connect to providers

To manage and monitor your agents, it is necessary to log into your respective providers. While the Agent Control Tower is accessible to anyone with Agentstudio access, each provider has its own prerequisites. To learn more about providers, refer to Getting started with providers.

AI-catalog your agents

Using the AI-catalog feature in custom providers you can add your agents using different types of files. The accepted file types include PDF, DOCX, YML, XML, JPG and PNG. This feature makes your agent addition process quick and seamless. For more information, refer to AI- cataloging your agents.

View and manage your agents

You can view and manage your agents all together or through their provider accounts. You can add trust levels and tags for easy filtering and organisation. Some providers such as Amazon Bedrock allows you to enable and disable an agent right from the Agent Control Tower. For more information, refer to Viewing and managing your agents.

Monitor your agents

Through the monitoring dashboard you can keep track on the number of times an agent is invoked, the average time for a process, spot errors etc. For more information, refer to Monitoring your agents.

Resources

On this Page