search
Ctrlk
Get support

Example - Setting first Make MCP Tool

To illustrate how to prepare a Make scenario for use with MCP, let’s walk through an example. This example will be relatively simple, but it demonstrates the required components (inputs, outputs, descriptions):

The “Greeting Generator” scenario is a simple Make automation designed to produce a personalized greeting message based on a provided name. It takes a single input, Name (Text), which represents the person to greet. The scenario processes this input using Make modules—such as a text aggregator or set variable module—to create a friendly message like “Hello, John! Welcome to our service.”

The resulting message is returned through a Return output module as a single output field: GreetingMessage (Text). While this example keeps the logic minimal for demonstration purposes, the scenario could easily be expanded to support additional features.

To set this up in Make:

1

hashtag
Create a new Scenario

Create the scenario (if not already) and set Scheduling to On demand (so it only runs when called). On the canvas, design the flow to produce the desired output. In our example, we might use a simple module (like Text Aggregator or Set Variable) to compose the greeting string using the Name input.

2

hashtag
Define Scenario Inputs

Click the Scenario Inputs icon (above the canvas) and Add item for each input. In our case, add an input:

  • Name: "Name"

  • Type: Text (string)

  • Description: "The name of the person to greet."

    (If you had more inputs, add each with its type and description accordingly.)

3

hashtag
Add Return Output Module

Add a Return Output module: At the end of your scenario’s flow, add the module Scenarios > Return output. When you drop this module, it will prompt you to define the output fields (if not already defined). Add an output field here:

  • Name: "GreetingMessage"

  • Type: Text

  • Description: "A personalized greeting generated by the scenario."

    After adding this field, map the actual greeting text into this field in the module (e.g. map the result of your text aggregator into the GreetingMessage field of the Return module). This ensures whatever greeting the scenario creates is captured as output.

4

hashtag
Add Scenario Description

Go to Diagram tab > Options > Edit description. Enter a concise description like “Returns a greeting message for a given name. This scenario takes a Name as input and outputs a friendly greeting containing that name.” Then save. This description will be visible to the AI as the tool’s description.

5

hashtag
Test

If you want to test it manually, you can run it from the Make interface, providing a sample name in the inputs dialog, and see that it returns the GreetingMessage in the output. You can also test via the Make API (using the Run scenario endpoint) to ensure it returns the output in the API response – this is exactly what the MCP server will do behind the scenes.

Once the above is done, your scenario is MCP-ready. When your Make MCP server (local or remote) is running, it will detect the “Greeting Generator” scenario. The AI assistant will see a tool named “Greeting Generator” (often under a category for your MCP server, e.g. make).

The tool will show that it requires a Name (with the description you wrote) and that it returns a GreetingMessage. If you ask the AI to use this tool – for example: “Use the Greeting Generator to greet John” – the assistant will supply Name = "John" to the MCP server, which triggers the scenario. The scenario runs, produces “Hello, John! …” as the GreetingMessage, and the MCP server returns that result. The AI then receives the output and can incorporate it into its response (e.g. telling you the greeting or performing whatever follow-up action is appropriate).