In the following use case, we will build an agent that helps us manage the stock inventory of a small shop. 

We will use Slack to submit tasks for the agent and receive responses about their completion. The agent will use 

 and other information we provide to him to list our stock information and order new items.

API key of the AI service provider. Check the documentation of your AI service provider to find out how to get it.

Build your first AI agent step by step with this simple inventory management example

Make AI Agents app

AI agent use case

A safety feature that saves unfinished scenario runs when errors occur, preventing data loss. Stored in the Incomplete Executions tab, you can retry automatically, resolve manually, or delete them. Disabled by default, enable it in Scenario settings.

Incomplete_executions

incomplete_executions

Error handlers act as a safety net that catches unexpected failures to prevent your scenarios from failing. They allow you to define exactly how the system should react, such as by retrying the task, ignoring the mistake, or rolling back changes.

error_handler

The Rollback error handler stops a scenario and reverts changes made by modules that support transactions to ensure data remains consistent when an error occurs.

Rollback

An automated workflow within the Make platform. You build it by connecting modules together, and it runs based on a trigger you set up.

scenario

a list or sequence of characters (like letters, numbers, and symbols) used to represent text in a computer programming.

string

An operation is a single module run to process data or check for new data. 

Operations

Credits (formerly operations) are the currency for running scenarios.

Credits

A route is a path that data can flow through in a scenario, visualized as the dotted lines between modules.

Route

scenario_plural_lowercase

scenario_singular_lowercase

Scenario_plural

Scenario_singular

product_name

Step 1. Set up the AI agent

Help Center

What is Make?

What's an API?

Learn the basics

Step 1. Plan your scenario

Step 2. Get your apps ready

Step 3. Add your first app

Step 4. Create a connection

Step 5. Set up the trigger

Step 6. Test the module

Step 7. Add another module

Step 8. Map data

Step 9. Test your scenario

Step 10. Schedule your scenario

Create your first scenario

Step 1. Get your app ready

Step 2. Add a router

Step 3. Set up another module

Step 4. Add a filter

Step 5. Test your scenario

Step 6. Add an aggregator

Step 7. Test the final scenario

Expand your scenario

Get started

Credits & operations

Connect an application

Schedule a scenario

Clone a scenario

Active and inactive scenarios

Scenario execution flow

Scenario history

Scenario blueprints

Scenario templates

Scenario sharing

Scenarios & connections

Introduction to Make apps

Types of modules

Module settings

Update legacy modules with new modules

Webhooks

Select the first bundle to process

Apps & modules

Item data types

Type coercion

Mapping

Filtering

Mapping arrays

Working with files

Data & mapping

Aggregator

Iterator

Router

Tools

Keyboard shortcuts

Resources

Key concepts

Scenario settings

Scenario notes

Scenario inputs and outputs

Subscenarios

Create and manage scenario templates

Custom scenario properties

Variables

Incomplete executions

Scenario execution, cycles, and phases

Restore and recover scenario

Scenario run replay

Scenarios

Call Google APIs with OAuth 2.0 HTTP request

Connect to any web service using OAuth 2.0

Connect to Google services using a custom OAuth client

Replace connections across multiple modules

Credential requests

Dynamic connections

On-premise agent

Devices

Keys and certificates

Allow connections to and from Make IP addresses

Connections

If-else and Merge

Flow control

Converger

Text parser

Use functions

General functions

Math functions

Math variables

Text and binary functions

Date and time functions

Tokens for date/time parsing

Tokens for date/time formatting

Array functions (CXPDOC2-13716)

Array functions

Custom functions

Functions

Methods of securing data

AES (Advanced Encryption Standard)

PGP (Pretty Good Privacy)

Digital signature

Hash functions

Securing data with Make

Data stores

Data structures

Make DevTool

Developers

Explore more

Introduction to Make AI Agents (New)

Create your first AI agent

Sales outreach AI agent use case

Slack-triggered AI agent

Telegram-triggered AI agent

Email-triggered AI agent

Form-triggered AI agent

Webhook-triggered AI agent

Mailhook-triggered AI agent

Create AI agents for different triggers

Knowledge

Make AI Agents (New) app

Make AI Agents (New) best practices

Make AI Agents (New)

Introduction to AI agents

Manage AI agents

Module tools for AI agents

Scenarios for AI agents

MCP tools for AI agents

Tools for AI agents

AI agent best practices

AI Agents configuration

Step 2. Create the AI agent's tools

Step 3. Create a scenario to send tasks to the AI agent

Make AI Agents

Introduction to MCP

Get started with Make MCP server

Make MCP server

Connect to Claude Desktop

Connect to ChatGPT

Connect to Cursor

Connect to OpenAI

Connect to OpenAI API

MCP toolboxes

Model Context Protocol

Introduction to Make Grid

Access Make Grid

Explore Make Grid

How to use Make Grid

Tutorial: How to replace a dependency

Best practices to use Make Grid

Make Grid

Introduction to errors and warnings

Overview of error handling

Fix rate limit errors

Fix connection errors

Fix missing data errors

Fix general errors

How to handle errors

Quick error handling reference

Break error handler

Commit error handler

Ignore error handler

Resume error handler

Rollback error handler

Error handlers

Types of errors

Types of warnings

Exponential backoff

Throw

Error handling

Organizations

Teams

Organizations & teams

Change your subscription

Subscription

Your organization