Can lightning’s revocation mechanism breaks k-of-n security?

[hmbdofficial]

I saw this proposal on a news outlet which I began to research on, and I found the Shachain part of the k-of -n lightning proposal really interesting. Thus implies that with ordinary Multisig, we can clearly say that an attacker need to compromise k singers before they can spend the funds. But from my research I understand that lightning also depends revocation secrets to protect an old channel state being broadcast.
Now my question is could it be that if an attacker can obtain the revocation secrets without compromising k singers prove the insecurity of the lightning node k-of -n? Or does true k-of-n security require the revocation mechanism itself to have the same threshold security as the signing keys?
Source:
https://delvingbitcoin.org/t/towards-a-k-of-n-lightning-network-node/2395

[Thawson]

From what I understood, that's the problem that the proposal is supposed to solve.
Protection of the funding keys is already possible by k-of-n, but the revocation part is not necessarily the same. Why does it add multiple shachains, to make it so that an attacker would still have to compromise at least k signers before it can be able expoit the revocation mechanism?