We made some changes to the API and made one single endpoint that will let you control what kind of respons you want back
Please see this example for

• https://jsfiddle.net/fjqjhpha/ - How to use the old css query selector
• https://jsfiddle.net/eho87va7/ - How to take screenshots
• https://jsfiddle.net/6met4xs7/ - How to use FormData to take multiple screenshots
• https://jsfiddle.net/3b8223be/ - And play with a empty page and use canvas element to draw things

(The demos heavily rely on ES6 fetch, template string and arrow function and may not work in all browser)

The special javascript command is

• screenCapture() to take a screenshot of how the page currently look like and get back a Blob file (image/png)
• respond.send(...) Can take anything the window.fetch api support as body (can be - Blob, FormData, string, json, arrayBuffer)
• resizeTo(x, y) This exist natively in the browser but will help you take different responsive screenshots

This new api makes it possible to receive anything you throw in to respond.send() it can also work Async so you can use setTimeout, wait for any picture to load and so on.
Now you have more control over how long time it will take to pre-render