Loading
One number for how much can go wrong, built from risk alone and never from return. Higher is worse.Note 13
Read from the contract, never declared. A single private key and a multisig that requires one signature are the same thing and are counted together. Capital whose control we have not read is its own line, because unread is not unowned.
The score above says how much can go wrong. This says what follows from it for an agent spending money it cannot get back, under the Conservative policy the API applies: it will not commit capital to a band worse than moderate, nor where control is thin, which means we looked and found little standing between one party and the money.
The largest block of capital here fails on control, not on risk.
Band inside the policy and control read as checked.
We read the control and one party can move the money.
We read the band as high or severe.
Our gap, not a finding. The policy abstains.
Too little measured to score at all.
The two greys are not refusals. A vault whose control or band we never read is our silence, and the policy answers abstain for it, which is not an objection and not a pass. Folding those into one “would not allocate” figure would turn a gap in our reading into a verdict about somebody else, which is the thing this site exists not to do. Of $11.68B across 2,633 open vaults, $959.31M is in vaults this policy would act on.