Robust and versatile headless monkey testing for the modern web with reproducible steps, error alerts, strategy sharing and many other good things.
It's important to remember that monkey testing should be used in conjunction with smarter tests such as integration tests.
Once webmonkey has been installed globally, you can begin testing by supplying the --url parameter you'd like to test. As that's the only required field, testing begins immediately and proceeds with 50 actions.
foo@bar:~$ webmonkey --url https://news.bbc.co.uk/For other parameters you can type webmonkey --help at any time.
By default each action has an equal chance of being chosen at random to be run. However you can tweak this by passing the --strategy parameter which allows for a comma separated list of actions followed by how likely they are to be chosen โ for example --strategy clicker=10,reloader=0 would make clicker 10 times as likely to run than any other action โ aside from reloader which will never run.
Oftentimes you'll want to authenticate before proceeding with the testing. In cases such as these webmonkey provides a hooks file where you export two optional functions โ create and destroy โ you can specify the location of the hooks file with the --hooks parameter.
The hooks file must be in the *.mjs format โ for instance to authenticate on a fictitious website one might implement the following.
export const create = async (page) => {
await page.goto('https://www.example.com/');
await page.focus('#username');
await page.keyboard.type('webmonkey');
await page.focus('#password');
await page.keyboard.type('monkeynuts');
await page.keyboard.press('Enter');
await page.waitForNavigation({ waitUntil: 'networkidle0' });
};Currently we have the following set of monkeys that perform various actions on your supplied domain:

