That's why I ask about the non radio (actual atomic) clocks.
GPS satellites are in roughly 20 000 km orbits, so latency is at least 65 ms (more for slant range). Furthermore, while I'm pretty sure GPS sats keep correct time, the position calculation doesn't depend on the receiver getting the correct time, just that the clocks on the satellites are synced, since it's the time difference that matters for GPS.
Since GPS receivers don't have a two way communication capability, adjusting for this 65+ ms latency to get the correct time is not possible.
Frequency standards again don't need to keep time, just the beat, right?
That's why I ask if real, full blown atomic clocks are available to mere mortals. I honestly don't know and don't really have an idea for a ballpark figure.