I was able to implement everything I needed to control my HomeAssistant (via Node-red interpreting the HTTP GET calls) by using a different query for each function added to the IP control device. so to trigger the room for 'Movie Mode' as set in my automation flow I send a GET to "http://localip:1880/endpoint/remote?moviemode=true" and a function takes that query and sends off the required triggers to do that.
Then I have a single command for "all off" that is the only active power command configured for all activities, as all my source devices stay on and only the TV/Receiver/subwoofer etc get turned on and off. so when I switch activities only the source changes for example, but if I press the physical OFF key on the remote it sends a command "?turnoff=true" which runs a shutdown sequence.
So far this is working well and I will be adding controls and tinkering with it for some time, I have also used this to remap the mostly unused colour buttons on every activity to control the lighting and AC with short presses turning them up or down, and long press turning them on and off entirely.
This way i don't need to pass anything in the body text, though it would be good to know what variables we can send and whether its possible to expand on this functionality to automatically generate a set of commands so we dont have to cut/paste dozens of URLs and AUTH headers for each entry. a return API would be good too, possibly to update the name of things on the display with a variable or to add a dummy command into the current activity shortcut list that shows the current temperature for example.
So what's missing? well I'd like more images to choose from on the display for the activities, or to select colours for them. I'd like more control over macro timing, as the shortest delay time of 0.5s is too long for things like entering passcodes, but it becomes unreliable when set to zero delay, a 0.2s would be perfect.