Mendix Testing Kit
Being builtSets up Playwright testing in a Mendix app, gives your AI coding assistant the rules it should follow, and fills in what it can from your live model.
Why it exists
Passing the error pane doesn't mean a Mendix change works. Most real bugs get straight past the checker, so the only proof is the running app. Everyone knows this, and teams still ship on a green model, because proving it properly needs a test harness nobody has time to build.
And an assistant that hasn't been given the rules will make them up. It'll hardcode a record that moves after the next database rebuild, read an input the wrong way and report a bug that isn't there, or hand you a test document nobody ever runs.
What it does
The kit puts the harness and the rules in place in one click, so you've got both from day one. It works with any AI coding assistant, because the instructions are plain markdown, and it never overwrites a file you already have.
- A Playwright project that's ready to go. One project per role, saved sessions, helpers for Mendix selectors and sessions, a smoke test, and a script that waits for the app to really start.
- Instructions for your assistant, one file for each moment: finishing a change, running the suite, writing a test, checking before a run, and writing a manual test case.
- Facts from your live model. Modules, roles, every page with who can open it and its widget names, and every flow with who can call it.
- A panel in Studio Pro. Setup shows what's installed and adds what's missing. Checks lines up the steps only a person can judge, one at a time, with Pass, Fail and a note, and hands your answers back to the assistant.
How to use it
- Install the extension in your app. You can build it straight in or import it as an add-on module.
- In Studio Pro, go to Extensions › Testing Kit › Set up testing in this app.
- Then run Refresh model facts from the same menu, and point your assistant at the instructions it's installed.