While I think ideally, you would want some invoke-able agent command-line option, like the options here, I don’t see this as a currently available option. You might want to post this as a feature request.

So the only way I can think of how to do this would be to invoke the API’s immediate remediation endpoint to invoke the action directly on your client (ensuring you’re not storing any api tokens in plain text).

Then you could monitor the output of the worklet by making a secondary API call to gather device details. If the status indicated a success, you could proceed with whatever other actions you want. Rinse and repeat.

Kinda complicated, but doable!