What is the least computationally intensive way to identify Patescibacteria (CPR)?

I’m writing a wrapper that runs CheckM CPR workflow on Patescibacteria and I run GTDB-Tk first. GTDB-Tk is quite overkill for the purpose of what I’m trying to do so I’m wondering if there are is a way I can adjust the parameters or workflow to basically say “this genome is Patescibacteria or this genome is NOT Patescibacteria”. Does anyone have any recommendations?