Does using an EEM script to monitor high CPU of Cisco Switches/Routers cause additional overhead?

So I want to pro-actively monitor my customers network. I am thinking of deploying the below EEM script. However, I want to know if the below script would add any additional overhead to the device resources? (Assuming that the script is going to poll the CPU details in very short intervals)

Link to EEM Script

Also, if there are any alternative ways of doing this, please suggest.

Customer limitation: We cannot deploy a monitoring tool in their environment. That's out of the picture.