Skip to main content
Feedback

Agent step

Updated 

The Agent step allows you to incorporate AI agents built with Agentstudio into your integration processes. This allows you to add non-deterministic actions to processes to unlock new automations for your business. For example, you can schedule agents to regularly receive invoice PDF files from your financial applications, read them, summarize their details, and send notifications via email about discrepancies.

Find the Agent step in the Execute section of the Steps Palette. After adding it to the canvas, you can choose from the available agents in your Control Tower to configure in the step.

To configure and use the Agent step, refer to Configuring the Agent step. For agentic process design patterns, refer to Integration process design patterns. For known issues and common questions, refer to Agent step troubleshooting and the Agent step FAQ.

Agentic use cases for the Agent step

Refer to Agentic use cases for customer success stories and common use cases for agentic workflows.

To help you identify integrations in your account that are ready to use with the Agent step, refer to Installing recommended agents to learn more about agent templates available in Agentstudio.

Prerequisites

  • You must have Agentstudio to configure the Agent step. Contact your Boomi representative to add this feature.
  • You must have an online integration runtime to browse and select agents in the Agent step. It is not mandatory to have a public integration cloud attachment to use the Agent step.
  • If using a self-hosted runtime to browse available agents, you must add IP addresses for communicating with Agent Control Tower to your allowlist. For more information, refer to Hostnames and IP addresses for the Boomi Enterprise Platform. Self-hosted runtime support is a part of the Flex Runtime Early Access program.

Important considerations

  • You can configure the Agent step with any agents created in Agentstudio. External agents, even those managed by Agent Control Tower, are not currently available for use with the Agent step.
  • The Agent step accepts a single input and produces a single output, so agents with multi-turn response, complex API tools, or chatbots are not compatible.
  • Context window limitations: User prompt, agent goals, tasks, and instructions combined cannot exceed 200K input tokens.
  • The Agent Designer supports XML and JSON API responses.
  • Agentstudio truncates tool responses greater than 100k tokens.
  • Regenerate configurations in the Agent step after you change the agent type in Agentstudio to reflect the updated agent definition.
  • Be cautious when updating agents in a deployed process, as these changes can cause issues or process failures.
  • You cannot create Agent connections or operations through the Component API.
  • When you deploy a new version of an agent, existing Agent step configurations remain unchanged to ensure stability. Sync your agents in Agent Control Tower, then manually select the new version in step configuration to accept agent updates in your process.
  • Since multiple packaged versions share the same deployment_id, an Agent step executes the latest deployed version (v2) even if the UI still displays v1. However, to avoid execution failures caused by schema mismatches or invalid configurations from deleted agents, you must manually sync to ensure the Agent step reflects the current active schema.

Agent response modes

When building an agent, you can choose how an AI agent responds in the Profile screen in the Agent Designer. There are two response modes.

Structured mode is ideal for agentic workflows in Integration.

  • Structured mode - The AI agent receives JSON input from Boomi Integration and responds in a consistent, structured JSON format. This mode is ideal for workflows where the agent's response is consumed by downstream systems. The response is single-turn, meaning the agent has a single input and output with no reference to previous responses.
  • Conversational mode (default) - The AI agent responds in natural language text. This mode is ideal for conversational agents where the user may have follow-up questions after the initial prompt. The agent can use chat history and context to respond to the user and achieve its goal.

The mode you choose determines your Integration design pattern. Refer to Integration process design patterns for agentic workflows.

The following video provides a quick overview of using structured agents in the Agent step:

Agent deployments

You can select agents deployed to regional Boomi-managed runtime clouds, such as ANZ, GBR, and Japan, directly within the Agent step. For more information, refer to Deploying agents.

The Bring Your Own Model Early Access feature in Agentstudio allows you to deploy your agents to self-hosted runtimes. Refer to Using your own models with agents for more information.

Browsing available agents in the Agent step

The Agent step includes a runtime selector above the agent list. Use it to browse and select agents from any runtime you have role-based access to, including runtimes other than the account's public cloud.

  • In accounts with access to multiple public cloud runtimes, if access to one is removed, the Agent step automatically selects another runtime you still have access to.
  • The runtime selector only lists runtimes you have role-based access to through Runtime Management.
note

You need to select a runtime again from the selection dialog when any of the following occurs:

  • An error occurs while building the connection.
  • Agentstudio syncs.
  • You regenerate the configuration.

You select a runtime once per Agent Panel instance. That selection applies only to the current instance and doesn't carry over to the account or session.

Selecting an agent in the Agent step

The Agent step displays available agents with the following columns: Agent, Runtime, Version, and Account. Hover over any agent row to view a tooltip showing the agent's name, version, runtime, account name, and ID. Use this information to identify and select the correct agent when the same agent is deployed across multiple regions or versions.

Select the region to filter and display only agents associated with that specific region. This is useful when your account has agents across multiple regions. You can also search for a specific agent using the Search Agents search bar. To view the most recent list of agents from the Agent Control Tower, select Refresh.

  • Only the latest active deployed version of an Agent to the Runtime is displayed for selection under the Agent step.
  • For existing agents which were active before May 2, 2026, the default region and runtime are considered US and version 1.0.
  • Self-hosted runtimes are categorized as a "Self-Hosted" region. Refer to Flex Runtime for information on this Early Access feature.

Next steps

Configure the Agent step in your integration process. Refer to Configuring the Agent step.

Additional resources

On this Page