Generative MUnit Tests With MuleSoft Vibes

Reading time: 9 minutes

Testing is critical to integration development, but let’s be honest: writing MUnit tests can feel like a chore. You need to account for happy paths, error scenarios, edge cases, and realistic mock data. It’s time-consuming work that often gets deprioritized.

MuleSoft Vibes: Generative MUnit Tests changes that. This feature uses generative AI to create complete MUnit test suites directly from your Mule flows, including mock data that mirrors your actual payloads.

Prerequisites

• Anypoint Code Builder (Desktop or Cloud IDE) with MuleSoft Vibes enabled
• An existing Mule project with at least one flow to test

Steps for getting started

For this tutorial, we’ll use a simple login flow that accepts a POST request with username and password. If credentials match (mule/mule), it returns a success HTML page; otherwise, it returns a failure HTML page. This flow has two distinct paths (successful and unsuccessful login), making it a perfect candidate for unit testing.

Step 1: Open your project and launch vibes

Open your Mule project in Anypoint Code Builder. Click the Mule vibes icon in the Activity Bar to open the AI ​​assistant panel.

Step 2: Generate the MUnit test suite

Now, let’s ask Vibes to generate tests for our login flow. Enter the following promptly in the Vibes chat:

Generate MUnit tests for the DoLoginFlow in login.xml. Include tests for both successful and unsuccessful login scenarios. Generate appropriate mock data for each test case.

Mule Vibes wants:

• Analyze the login.xml file to understand the flow structure
• Identify the Choice router and its two branches
• Generate a test suite file (eg, login-test-suite.xml) in src/test/munit
• Create mock data files (payload and attributes) in src/test/resource

Step 3: Review the generated test suite

Once Vibes completes, navigate to src/test/munit/ and open login-test-suite.xml. You should see two tests:

• DoLoginFlow test (Successful login scenario)
• DoLoginFlow-unsuccessful (Failed login scenario)

Each generated test follows the standard MUnit pattern:

Section	Purpose
munit:behavior	Sets up the input event (payload and attributes) using the generated mock data
munit:execution	Invokes the flow under test via flow-ref
munit:validation	Asserts the expected output and verifies that the correct branch was executed

Successful login test example

Step 4: Examine the generated mock data

Vibes also generates realistic mock data based on your flow’s metadata. Navigate to src/test/resources/ to find:

• DoLoginFlowtest/set-event_payload.dwl – Payload for successful login
• DoLoginFlowtest/set-event_attributes.dwl – HTTP attributes (headers, method, etc.)
• DoLoginFlowunsuccessful/set-event_payload.dwl – Payload for failed login

Successful login payload example

{
 "username": "mule",
 "password": "mule",
 "Submit": "Submit"
}

Unsuccessful login payload example

{
 "username": "invalid_user",
 "password": "wrong_password",
 "Submit": "Submit"
}

Step 5: Run the tests

Right-click on login-test-suite.xml in the Explorer. Select Run MUnit tests. Then, observe the results in the MUnit panel. Both tests should pass, confirming that the unsuccessful login path returns the failure HTML, and the successful login path returns the expected HTML.

Step 6: Add a new test scenario

What if you want to test a specific edge case? Simply ask Vibes to add it.

Add a new MUnit test for DoLoginFlow that tests when the username is empty. The test should verify that the login fails.

Mule Vibes will add a new test case to the existing test suite, generate mock data with an empty username field, and configure the validation to assert the failure path.

Step 7: Modify an existing test

You can also ask Vibes to update tests as your flows evolve.

Update the DoLoginFlow-unsuccessful test to also verify that a logger was called with the message "User is trying to login".

Vibes will modify the section to add a verify-call assertion for the Logger component.

feature	Benefit
Generate a complete test suite	Analyzes flow logic and creates tests for all paths
Create realistic mock data	Uses application metadata to generate accurate payloads
Add new test scenarios	Responds to natural language requests for edge cases
Modify existing tests	Updates tests as your flows evolve

Streamline your testing today