How to Use AI Agents in Make.com

This step-by-step guide explains how to set up, configure, and use AI Agents in make.com so you can build powerful, conversational automations without writing code.

AI Agents are a beta feature that let you create reusable, chat-based assistants that talk to users, gather information, and trigger scenarios. They combine AI Workbooks, data sources, and structured outputs to help teams automate complex tasks.

What Are AI Agents in Make.com?

AI Agents in make.com are reusable AI-powered assistants that can:

Engage in multi-turn conversations with users
Collect and structure information from chat
Trigger make.com scenarios and workflows
Use custom knowledge sources or tools

Each AI Agent is defined by three building blocks:

Agent setup – the name, description, and routing keys
Core AI behavior – the AI Workbook that defines logic and tools
Goal and output structure – how the Agent formats and returns data

Once created, these Agents can be reused across multiple entry points, such as forms, chat widgets, or integrations you connect to make.com.

Prerequisites for Using AI Agents in Make.com

Before building an AI Agent, make sure you have:

An active make.com account with access to AI features
At least one AI Workbook (or the ability to create one)
Understanding of your use case and the data you want the Agent to collect

AI Agents are built on top of AI Workbooks. The Workbook defines the tools, instructions, and logic the Agent will follow during a conversation.

How to Create a New AI Agent in Make.com

Follow these steps to create a new AI Agent in make.com from the AI Agents homepage.

Step 1: Open the AI Agents Page in Make.com

Log in to your make.com dashboard.
In the left navigation, locate the AI or AI Agents section.
Open the AI Agents homepage where existing Agents are listed.

This page is the central place to manage, monitor, and edit all of your AI Agents.

Step 2: Start a New AI Agent

On the AI Agents homepage, click the button to create a new Agent.
Provide a name that reflects the Agent’s purpose, such as “Support Triage Agent”.
Add a description so teammates know what this make.com Agent does.

The name and description help you organize multiple Agents and locate them when connecting to different workflows.

Step 3: Configure Basic Agent Settings

In the Agent configuration view, define how the AI Agent behaves and how it will be invoked in make.com.

Agent name – used throughout your workspace.
Description – explains the Agent’s role and scope.
Tags or groups (if available) – organize Agents by department or use case.

These settings are foundational and can be updated as your automations evolve.

Connect an AI Workbook to Your Make.com Agent

Every AI Agent in make.com relies on an AI Workbook to define its logic, tools, and knowledge. Connecting a Workbook is essential for the Agent to respond meaningfully.

Step 4: Choose or Create an AI Workbook

In the Agent configuration panel, locate the option to attach an AI Workbook.
Select an existing Workbook from the list, or create a new one tailored to this Agent.
Confirm the connection so the Agent can use the Workbook as its brain.

An AI Workbook can contain tools, custom instructions, data sources, and structured outputs that shape how the Agent behaves in make.com conversations.

Step 5: Define the Agent’s Goal and Output

Inside the Workbook linked to your Agent, define:

Goal – what the Agent is trying to achieve (for example, “collect user problem details and prioritize urgent cases”).
Desired output structure – the fields or schema the Agent should fill, such as name, email, issue category, and urgency level.

This structured output can then be passed directly into make.com scenarios, forms, or external systems for further processing.

Testing and Debugging AI Agents in Make.com

Testing is crucial to ensure your AI Agent behaves as expected before you expose it to end users.

Step 6: Use the Built-in Test Console

Open the specific Agent you want to test in make.com.
Look for a built-in chat or test console on the Agent detail page.
Send sample messages that mirror real user questions or requests.

Observe how the Agent responds, what data it gathers, and how it structures outputs. Adjust the Workbook instructions or tools where necessary.

Step 7: Tweak Prompts, Tools, and Knowledge

If your Agent’s replies are incomplete or inaccurate, refine:

System or agent instructions in the Workbook
Tools and functions available to the Agent
Knowledge sources, such as documents or FAQs
Output schema to ensure essential details are always captured

Repeat testing until the AI Agent consistently performs the way you expect inside make.com.

Connecting AI Agents to Make.com Scenarios

Once your Agent is stable, you can connect it to scenarios to trigger downstream automation.

Step 8: Use AI Agent Outputs in Scenarios

Create or open a scenario in make.com.
Add modules or webhooks that receive the Agent’s structured output.
Map fields from the AI Agent to other apps or services in the scenario.

Typical follow-up actions include creating support tickets, logging CRM records, sending notifications, or launching additional AI flows.

Step 9: Reuse the Same Agent Across Channels

One AI Agent in make.com can power multiple channels, such as:

Web chat widgets
Support intake forms
Internal helpdesk bots
Custom integrations built via API

This reuse keeps behavior consistent and centralizes updates in a single Agent and Workbook.

Monitoring and Improving Your Make.com AI Agents

After deployment, continuously monitor performance and refine configuration to improve accuracy and user satisfaction.

Step 10: Review Conversations and Metrics

Regularly review transcripts where your make.com Agent struggled.
Identify missing instructions, tools, or data sources.
Adjust the Workbook and test again.

Iterative improvement helps your Agent become more reliable over time.

Additional Resources for Make.com AI Agents

To deepen your understanding of AI Agents in make.com and see the full official documentation, review the guide at Introduction to AI Agents. It covers the latest capabilities and interface details.

If you need expert help designing scalable AI workflows, prompt strategies, and complex automations around make.com, you can consult specialists at Consultevo for implementation support.

By following these steps, you can confidently build, test, and launch AI Agents in make.com that automate conversations, collect structured data, and orchestrate powerful workflows across your tech stack.

Need Help With Make.com?

If you want expert help building, automating, or scaling your Make scenarios, work with ConsultEvo — certified workflow and automation specialists.

Get Help

Getting Started with Make.com AI Agents