For a project we need to run a script when there are active triggers in a host group and run another script when the are no more triggers.

This is not easy: if I configure an action such as:

(A) Maintenance status not in "maintenance"
(B) Trigger value = "PROBLEM"
(C) Host group = "Linux Servers"

the script would be run for every single trigger in the host group and the recovery action is even harder...

A zabbix internal check such as: zabbix[triggers,<hostgroup>,<minseverity>] would be nice

For example zabbix[triggers,Linux Servers,high] should return the number of active triggers with at least high severity in the host group "Linux Servers".