Admin Keys and Multisig Control

How it happens

Most deployed protocols retain privileged roles: an owner that can change parameters, a guardian that can pause, a minter that can create tokens, an upgrader that can replace the logic entirely. These are usually held by a multisig requiring some threshold of signers, which reduces the risk of one person acting alone but concentrates the system into a handful of devices and people. Compromise of enough of those keys is equivalent to compromise of the protocol, without any flaw in the contract code, and the same is true of coercion or of a legal order directed at identifiable signers. A timelock changes the picture materially, because it turns a silent change into a publicly visible pending change with a window in which users can withdraw.

What you can actually observe

Enumerate the privileged roles in the deployed contracts and resolve each to an address, then check whether that address is an externally owned account, a multisig, or a timelock, and what the multisig threshold and signer count are. Check whether signers are publicly identified, whether they are independent of one another, and whether they use separate hardware and jurisdictions. Read what the pause and mint functions can actually do, since a pause that also blocks withdrawals is a different instrument from one that only stops deposits.

Precedent

The Ronin bridge was drained in 2022 after attackers gained control of a majority of the keys in its validator multisig, with no flaw in the contract logic involved.

What makes it more or less material

Weigh what the privileged roles can actually do, the threshold and independence of signers, whether a timelock delays changes, whether users can exit within that delay, and whether an emergency path bypasses it.