Migrating From mabl To KaneAI

This guide covers your mabl migration to KaneAI: what you can move to TestMu AI, and how.

[Note: KaneAI does not import or convert mabl tests. This guide follows a manual path. Export code from mabl and run it on TestMu AI, import test cases from CSV, and re-create tests in KaneAI in natural language.]

What Moves From mabl and Where It Goes

mabl asset	Where it goes on TestMu AI	How
Browser tests, as code	TestMu AI cloud grid	Export as Playwright or Selenium IDE, then run
Browser tests, as KaneAI tests	KaneAI	Re-create in natural language, or automate an imported test case
Tests exported as CSV	Test Manager test cases	CSV import
Flows	KaneAI modules	Re-create the steps as a module
Variables	KaneAI variables	Re-create them
DataTables	Datasets in Test Manager	Export as CSV, then use Import Data
API tests	API steps in KaneAI tests	Export as Postman JSON, then paste requests as curl commands
Mobile tests	KaneAI app tests	Re-create them
Run history and artifacts	Stay in mabl	Export and keep as records

Before You Start

A TestMu AI account with KaneAI access. The Automate with KaneAI action requires it. Create one on the TestMu AI signup page.
Your Username and Access Key, from Account Settings > Password & Security.
Node.js for the mabl CLI. mabl requires an actively supported Node.js LTS version, currently 22 or higher. It does not support running its CLI from Windows Subsystem for Linux (WSL).

Step 1: Export Your mabl Tests

mabl tests export exports a mabl test as Playwright (TypeScript), Selenium IDE, JSON, CSV, YAML or Postman. YAML is the default. The Playwright and Selenium IDE formats are only supported for browser tests.

Install the mabl CLI, log in, and list your tests. mabl auth login --auto opens the mabl login page in your browser. mabl tests list returns test IDs, 10 tests by default. Set --limit to list more.

Verified

npm install -g @mablhq/mabl-cli
mabl auth login --auto
mabl tests list

Export each test in the format you plan to run:

Verified

Playwright (TypeScript)
Selenium IDE (.side)

mabl tests export {test-id} --format playwright --environment-id {environment-id}

--environment-id generates selectors from the find information mabl learned in that environment. mabl environments list returns environment IDs.

mabl tests export {test-id} --format side

The file goes to your present working directory, named after the test ID, unless you set --file. If you plan to import test cases in Step 3, also export each test with --format csv.

Step 2: Run the Exported Tests on TestMu AI

These runs are not KaneAI tests. Their results appear on the TestMu AI Web Automation Dashboard. Some mabl step types are not supported in these exports, as listed in What Does Not Carry Over.

Set your credentials as environment variables:

Verified

macOS / Linux
Windows (CMD)

export LT_USERNAME="undefined"
export LT_ACCESS_KEY="undefined"

set LT_USERNAME=undefined
set LT_ACCESS_KEY=undefined

Verified

Playwright on TestMu AI
Selenium IDE on TestMu AI

mabl's docs do not describe the structure of the Playwright export. For TypeScript tests that use the Playwright test runner, see How to Run TypeScript Tests With Playwright. Its lambdatest-setup.ts fixture makes the TestMu AI CDP connection.

Where your test launches a local browser, connect to the TestMu AI CDP endpoint instead. The capabilities object selects the browser and OS. Sample from How to Migrate Existing Playwright Tests:

const { chromium } = require('playwright')
const { expect } = require('@playwright/test');

(async () => {
  const capabilities = {
    'browserName': 'Chrome', // Browsers allowed: `Chrome`, `MicrosoftEdge`, `pw-chromium`, `pw-firefox` and `pw-webkit`
    'browserVersion': 'latest',
    'LT:Options': {
      'platform': 'Windows 10',
      'build': 'Playwright Sample Build',
      'name': 'Playwright Sample Test',
      'user': process.env.LT_USERNAME,
      'accessKey': process.env.LT_ACCESS_KEY,
    }
  }

  const browser = await chromium.connect({
    wsEndpoint: `wss://cdp.lambdatest.com/playwright?capabilities=${encodeURIComponent(JSON.stringify(capabilities))}`
  })

  // Test code ...

  await browser.close()
})()

Install the SIDE Runner:

npm install -g selenium-side-runner

Run the exported file on the TestMu AI Selenium grid. Replace user-name, access-key and <exported-test>.side with your values:

selenium-side-runner -w 4 "<exported-test>.side" --server https://user-name:access-key@hub.lambdatest.com/wd/hub -c "browserName='chrome' browserVersion='latest' platformName='Windows 10'"

-w sets the number of parallel processes. -c sets the browser, version and platform. See Run Selenium IDE Tests.

Step 3: Import Test Cases Into Test Manager

Test Manager has no mabl importer, so use CSV import. mabl's export docs do not list the CSV columns, so open the file and check them first.

In your Test Manager project, go to Add Test Case > Import CSV.
Choose Test cases use a single row or Test cases use multiple rows, to match how the steps are laid out.
Map the CSV headers to Test Manager fields. Mapping at least one CSV header to Title is mandatory. Steps, Outcome, Folder, system fields and custom fields are optional. Leave other columns as Ignore Field.
Map the values, review the preview, and start the import.

Each file can be up to 10 MB. See Import Test Cases from CSV.

If the columns do not map well, attach the CSV or JSON export to the AI Test Case Generator instead. It writes new test cases from the file rather than importing the mabl tests.

Step 4: Re-create Your Tests in KaneAI

Start From a Test Case

Open a saved test case and click Automate with KaneAI in the Automate section. KaneAI opens a session pre-filled with the test case's steps, where you author and run the test. This works one test case at a time, and the test case must have steps or scenarios (Automate with KaneAI).

Start a New Test

On the KaneAI dashboard, click Author Browser Test and write each step in natural language (Author Your First Desktop Browser Test). For mabl mobile tests, click Author App Test, select the Android or iOS tab and add your app (Author Your First Mobile App Test).

Manual Interaction records your clicks, typing and scrolling as steps. It works in desktop web browsers, Android apps and iOS apps in portrait, but not in mobile browsers. Add verify or assert steps as written instructions.

Rebuild Flows, Variables and DataTables

Flows are reusable sets of steps in mabl. In KaneAI, select the steps in a paused session and click Create Module. Reuse it with Add Module from the slash command menu (Modules).
Variables: use Add a variable from the slash command menu, or write Set username as John in natural language (Variables in KaneAI).
DataTables: export each one with mabl datatables export {datatable-id} and the --format csv option. In Test Manager, open Datasets, use Import Data in the dataset editor, then verify the data and save (Datasets).

Re-create Unsupported Step Types

mabl's Selenium and Playwright exports do not support these step types. The KaneAI docs below cover each one:

mabl step	KaneAI doc
Select steps	Forms, Inputs & Data Entry
Drag and drop steps	Drag and Drop
JavaScript snippets	JavaScript Execution
API steps	API Testing
File uploads	Upload and Download Files
MFA authenticator (TOTP) steps	TOTP Authentication
Conditional steps (Selenium exports)	Conditional Logic
Accessibility checks	Accessibility Testing in a Classic report test run, desktop Chrome and Edge only (Advanced Configurations)

Re-create API Tests

mabl exports API tests as Postman collections in JSON, from the test details page or the API test editor. KaneAI documents curl input for API steps, not a Postman collection import. To move each request:

Import the mabl JSON file into Postman as a new collection.
Open the request, click Code, and choose cURL.
In a KaneAI test, add an API step through the slash command and paste the curl command. You can paste several curl commands to add them all (API Testing).

Generate Code

KaneAI generates code for tests saved as Ready, where every step has run and passed (Code Generation):

Classic: Selenium Python and Appium Python by default. Selenium Java and Playwright Python, C# and JavaScript are available on request.
New Experience: Playwright Python and Playwright JavaScript. New Experience is rolling out in phases.

Contact support to enable an on-request framework or New Experience.

Step 5: Run Your KaneAI Tests

In your Test Manager project, open Test Runs, click Create Test Run, and set Type to KaneAI Generated Test Cases.
Add test cases. Only test cases with generated code are available.
Assign team members and update configurations as required, then save the test run.
Click Run on HyperExecute, then Run Instances Now or Schedule Test Run.

See Execute Test Runs on HyperExecute. Device scope, per Configurations:

With the KaneAI toggle on, the available platforms are Desktop and Real Device.
Native app testing on Real Devices is available only for KaneAI-generated test runs.
Virtual Mobile configurations are available for non-KaneAI test runs only.

For CI tools such as GitHub Actions or Jenkins, copy the Test Run ID from the test run URL. Send it as test_run_id in a POST request to https://test-manager-api.lambdatest.com/api/atm/v1/hyperexecute, with Basic Authentication (Test Automation with CI/CD).

What Does Not Carry Over

The tests. KaneAI does not import or convert mabl tests, and Test Manager has no mabl importer. KaneAI imports tests only from Katalon, for desktop web tests (Migration from Katalon). Test Manager's one-click importers cover TestRail, qTest, Xray and Zephyr Scale.
mabl-generated tests. Only user-generated tests export. The visit home page, link crawler and visual smoke tests do not.
Some steps. Select steps, drag and drop steps, JavaScript snippets, API steps, cookie interactions and file uploads are not supported in Selenium or Playwright exports. The same applies to download assertions, accessibility checks, MFA authenticator steps, email testing and PDF testing. Selenium exports also do not support conditional steps.
Code for other test types. Playwright and Selenium IDE exports cover browser tests only.
Run history and results. TestMu AI documents no importer for them. Before you leave mabl, export run artifacts with mabl test-runs export {run-id} --types all. This works for browser tests run in the mabl cloud. Analytics > Coverage can also download a CSV of passed and failed test totals by date, from the Test run history chart.