To my understanding, pools are only managed on the controller-level, whereas AutomateC is designed for targeting specific individual systems. If you want AutomateC and pooling to work, you need to toss out the built-in pool that Blue Prism provides and come up with your own internally-managed pool.
For example, we have a replacement piece of controller software that polls the bots in our ""pool"" to see which one can take on work. The next one available in that list is the one that is sent the command via AutomateC. Presumably, you could do the same using stages in Blue Prism, as long as it has a way of knowing which systems are available and which aren't. For this, you can run a query against the BPAResource table where AttributeID matches whichever resource group you have it configured for (these values should be listed in BPAResourceAttribute). So if you have a resource group named ""Pool"" (Note: Not a pool named pool), and BPAResourceAttribute shows that ID to be 3, you could querey against BPAResource where status = ready and AttributteID = 3.
If you go that route, I /highly/ recommend doing some organization of your machines. Otherwise, if every listed machine is in the same group, your code will start targeting dev machines as well. We've got separate groups called retired, local, debug and pool which we are able to discern pretty easily between.
As a side note, you can take it a step further and start getting creative with that particular kind of model. While a Pool (as defined internally by Blue Prism) will only let a single process run once within it, a custom pool (as outlined above) will let you run it as many times as you want. So if you have a queue with an unusually large work load, that can quickly get divided across multiple bots. You can then go a step further and configure each process to check if a stop was requested. Once that's in, you can halt the process on some (but not all) working bots and spin up another process that also needs to get worked. You could also add in a custom priority order and have it figure out which queue needs to get the most focus.
