In "proof of concept" is the concept proved or proofed?

It's certainly true that a "proof of concept" will not necessarily prove the concept. The test may fail. But that fact is hardly unique to this particular use of the word "proof". If you say, "I am looking for proof that the Polynesians discovered America", you may or may not find such proof. Etc etc.

When a proof of concept test is completed, if it was successful the concept is "proven" or "proved". If the proof of concept test failed, then the idea is either "disproven" -- we proved that it doesn't work -- or "unproven" -- the results were not adequate to say it whether it will work or not.

"Proof" as a verb is generally understood to refer to examination of text for errors, as in "The editor proofed the manuscript." I don't think I've ever heard someone say, "The engineer proofed the airplane's wings" or "The accountants proofed the tax software". Certainly not, "The physicist proofed his cold fusion theory." The verb "proof" has very little to do with the noun "proof".

When you have demonstrated a successful proof of a concept, it has been proved. After all, one of the definitions of prove is demonstrate the truth or existence of (something) by evidence or argument, as in the concept is difficult to prove.

Prove-as-in-test really only survives in stock phrases such as the exception that proves the rule : proof of concept seems too recent to have that meaning.

It's really a fixed, standalone phrase now, so I would avoid inflecting it : I'd just say a successful proof of concept has been completed.

The meaning of a proof of concept is not so much to test or examine an idea or technique for flaws as it is to persuade by demonstration that an idea or technique is useful.¹ So, given the choice, "prove" is better.