I too could make use of a Make Image type tool or Add-in.
Any guidance on how to set about making the Add-in? Transforms or Queries?
The bulk of the work is in writing a script that works just as a component.
Then you just take the script, save it as a file and put it into (a subfolder in) one of the Manifold folders (typically, ~\shared so that it is available in both 32-bit and 64-bit modes, and if you want to restrict it to just one mode, use either ~\bin or ~\bin64). The file then appears as a menu item under Tools - Add-ins. For running instances of 9, do Tools - Add-ins - Rescan to let the system refresh available add-ins after you save the file.
For the task at hand:
If we are talking about saving a portion of an image, I'd create a dialog asking for image name, the bounds to crop to and the scale. Then I'd generate a query that would compose a new image based on the input data, run that query, save the resulting image and delete it.
After that, we'll probably want some data from the UI - we can absolutely extend the object model to expose whatever is necessary.
Rendering drawings / labels / maps / layouts as images is quite a bit more involved and needs support from the side of the object model. We can absolutely add it, it's just a question of priorities.