The overall band reads low while 1 dimension read worse: controller high.
We checked 166 addresses: roles and direct holders against the digital currency addresses OFAC SDN published on 2026-11-13; none of them is on it.
Exact address matches against one list (OFAC SDN, version 55badef69280). Not a legal opinion, and no exact match is not a finding that no sanctioned party is involved.
Every daily reading we hold.
It quoted 21.92% at its highest and was still quoting on 2026-11-03. It quotes nothing now. A vault that has never paid and a vault that has stopped paying both read as nothing on the tiles above, and they are not the same proposition.
97.3% of the capital we can attribute here, or $1.35M, belongs to 120 addresses that also hold an open vault elsewhere, the deepest of them in 21. They can leave all of them on the same morning.
Pick or type an amount to weigh it against both: what the venue says could leave now, and the largest single withdrawal anybody has made here.
The ring marks the largest single withdrawal in the last 30 days, $22.0k.
The contract imposes no notice period and no request step: a redeem that succeeds settles in the transaction that asks for it, and it succeeds only if the vault holds enough to pay at that moment, which is the liquidity reading and a separate question.
There is no queue and no lock-up in the contract.
We have not established what could be taken out. We hold no reading of what could be taken out of this vault, and we have not asked the contract directly either. That is the absence of a reading and not a reading of nothing: it must not be read as the door being shut.
A quoted rate is one input, not a verdict.
It moves only when the strategy earns or loses; deposits do not move it.
Dashed: what the venue advertised, all recorded by us, which we began on 2026-09-19.
$10.0k would be $10.0k , a loss of $5.35, -0.1% over 85 days.
Priced from our first reading 5 days after that date; 2026-08-23, because it is the earliest price we hold rather than one we would have to invent.
Computed from the price of one share, so another depositor arriving is never counted as this strategy’s performance. Fees the venue charges inside the share price are included; anything charged outside it is not. It says what this vault did between two dates and nothing whatever about what it will do next.
Everything above is reconstructed from public data. A declaration is free, signed from this vault’s own key, and cannot be edited afterwards.Note 28
Six questions, and what we hold for each.
| Source mark | Question | What we hold | Source |
|---|---|---|---|
| What does it hold or lend into? | HXXIThe venue names a deposit asset. What the vault actually holds behind it has not been read. | Declared | |
| How much could leave now? | at least $1.81M idleOnly the unlent part is known, so this is a floor: at least this much can leave. Idle is not withdrawable, and a queue, a cooldown or a cap can stand between them. | Inferred | |
| Who runs it, and what else have they run? | D2 FinanceA named operator with a record across every vault they run, including the ones that closed. |
Scope: the protocol’s code; a curator’s configuration of it is not what these audits cover. Source: DefiLlama audit links, read 16 November 2026, 13:30 UTC.
the owner is a single externally-owned account, not a multisig
The owner is a single key rather than a contract, so there is no multisig or timelock behind it.
Locked here: every depositor we read with their share of the vault, the largest single wallet, and the ledger of each deposit and withdrawal. A vault that is eighty per cent one depositor is a vault whose price can be reset by a single withdrawal, and whose headline TVL says almost nothing about whether anyone else believes in it.
See what unlocksAlready a customer? Sign inWhat this page cannot show yet, and why.
One of those is not part of the band at all, so no amount of reading the band would have surfaced it.
Evidence · Not scored · 3 of 5 components read (55% of the weight)
Point at a risk to read what it rests on.
| Risk | Reads | In the band | From |
|---|---|---|---|
| Marketlimited scope: past share-price falls, markets shared with other vaults and, where the vault borrows, distance to liquidation; excludes collateral and borrower credit risk | Low | Yes | drawdown |
| Liquidity | Not read | Yes | nothing fed it |
| Depositor concentration | Low | Yes | concentration |
| Operatinglimited scope: continuity of the observed record; not a review of operations | Low | Yes | consistency |
| Controller | Highworse than the band | No | who may move the money |
| Evidence | Not scored | No | 3 of 5 risk-band components read (55% of the weight) |
The levels come from this vault’s own risk components and controller reading. The bars draw them; they are not a second calculation, and there is deliberately no overall figure.
Low observed risk is not zero risk. Market here is two observed readings: the worst fall in the vault's past share price and how much of its lending shares a market with other vaults we track. It does not assess the collateral, the borrowers' ability to repay, a concentration in one market (read under where the capital rests), or whether withdrawals hold under stress. The overall band weighs every component together instead, so it can read better than this and is not in conflict with it: a dimension says how bad its worst reading is, the band says what the whole weighs.
That is our coverage, not a finding.
Low observed risk is not zero risk. It measures how much of the vault its largest depositor holds, not how much the vault lends into one market; that is the allocation table. Input: largest wallet holds 19%, by the address's position; addresses are not merged by who owns them, and whether the largest is a person, another vault or a custodian is not established. The reading 0.00 is that share less 25%, over 75%: a quarter or less reads 0, the whole vault 1. The overall band weighs every component together instead, so it can read better than this and is not in conflict with it: a dimension says how bad its worst reading is, the band says what the whole weighs.
Low observed risk is not zero risk. The overall band weighs every component together instead, so it can read better than this and is not in conflict with it: a dimension says how bad its worst reading is, the band says what the whole weighs.
What that party may change is set by the venue's contract; if it can redirect funds and nothing delays it, one signature is enough to do so. This is not part of the overall band: a vault reading low can read high here.
By weight, 45 of the 82.5 points that apply were read, which is 55% of the component weight the band could draw on (a different denominator: components are not equally weighted); the rest of the 100 is the weight of what does not apply. This is a statement about our coverage and carries no level of its own, because how much we could read is not a risk the vault runs.
Every deposit and withdrawal drawn over the vault's value, the buckets of money in and out, the wallets that moved it and how big a whale is here.
See what unlocksAlready a customer? Sign inFigures computed from the share price read from the vault contract as we went: 2026-03-29 to 2026-11-16, 349 observations. 30-day window from 2026-10-17, 90-day window from 2026-08-18. Readings from the share price read from the contract at past blocks, reconstructed afterwards (52) exist and are not mixed in: one series, one source.
These are the same seven fields in the same order for every vault, so a high rate and a high rate nobody can withdraw from do not look alike.
| Measured |
| What does it depend on? | D2 Finance on HyperliquidThe venue and the chain are known; what the strategy routes into underneath them has not been read. | Declared |
| Who can change it, and how fast? | eoathe owner is a single externally-owned account, not a multisig | Measured |
| How much of the above was measured? | 2 measured · 2 declared · 1 inferredCounted from the rows above rather than asserted. Every figure on this page carries when it was read and what it was read from; the gaps are listed on /reconciliation with their size. | Measured |
This is not a score and there is no letter at the end of it: the answers are different kinds of thing and compressing them into one number would hide exactly the part a reader needs. Where we have nothing, the row says so.
The mark says where an answer came from, never whether it is good: a filled circle is a reading we made, a half circle an inference from readings we hold, quotation marks what the venue or operator states, and a dashed ring means we did not read it rather than that there is nothing to read.
Nothing here is a rating, a recommendation, or a statement that this vault is safe.
These fields are empty because nobody has attested them, not because we failed to find them.
An operator who registers gets a record that provably cannot be edited after the fact, and a grade no unregistered strategy can reach.
We have watched who controls this vault since 2026-08-24 and nothing we record has moved in that window.
That is not the same as a vault that has never changed hands: anything before 2026-08-24 predates this platform holding control state at all, and we do not reconstruct it from history.