An $825 Million Pre-Deposit Vault Filled in About 22 Minutes, and On-Chain Timestamps Show Most of It Landed Before the Public Was Told. Four Things to Check Before You Deposit Into Any Window.

The pre-deposit vault is, right now, the most common way a new chain or dollar comes to market. You put a stablecoin into an escrow before launch, you accrue points or a multiplier while it sits there, and you convert to the token at generation. Done honestly it is a reasonable instrument. It lets early capital signal intent and lets a protocol bootstrap without running a public sale. What went wrong last year is that the highest-profile version of it had a specific, documented failure, and that failure is easy to reproduce for anyone who copies the mechanics without the disclosures that would have caught it.

The case worth studying is Stable, the Bitfinex and Tether-linked Layer 1 that uses USDT for gas. Its Phase 1 pre-deposit vault opened on October 23 to 24, 2025 with an $825 million hard cap, and it filled in about 22 minutes. The fill time is real, but the part that matters is when the deposits actually arrived.

What the chain showed for Stable

According to on-chain analysis reported by The Block and The Defiant, the first deposits hit the vault contract roughly 22 to 25 minutes before the campaign was publicly announced. By the time the announcement went out, a cluster of about ten wallets, described in that reporting as linked to the vault owner, had already deposited around $500 million, close to 60 percent of the cap, and a single wallet accounted for more than half of the whole cap. Fewer than 300 wallets took part in Phase 1 at all, with reported counts running from 194 to 274. There were no per-wallet caps, no whitelist, and no announced start time, so for most people the window was open and most of the way full at the same moment they heard it existed. One commentator quoted in the coverage put the objection plainly: is a public sale really a public sale if you do not tell anyone until it is full.

A note on precision, because this is someone else's protocol. The link between those wallets and the vault owner is what on-chain analysts and reporters inferred from clustering and timing. Stable did not confirm it, and it did not respond to The Defiant's request for comment. The narrow version does not need the ownership question settled: the deposits are timestamped on-chain before the public announcement, the concentration is on-chain, and the team gave no explanation for either. That is enough to see the problem.

There was also no proof of reserves of the pre-deposited funds while the window was open, and at the time nobody treated one as expected. The only verifiability was raw Etherscan, which is how the pre-announcement deposits were spotted after the fact rather than blocked before it. Phase 2, in November, added the missing controls: KYC, per-wallet caps, and an announced start time. Then the front end run by Hourglass crashed under traffic and a denial-of-service attack, deposits were paused about three hours after opening, and users who had deposited directly through Etherscan had to be refunded. The person whose firm ran a competing pre-deposit summarized the lesson. For Plasma's earlier vault, Veda's cofounder said, they took great pains to hide the contracts until the public announcement, and deposits were blocked until then. So the countermeasure already exists in the market. Most launches simply do not use it.

The pattern is bigger than one launch

Stable is the clearest case of the pre-fill, but the surrounding failures show up across the category, and they are worth separating so this does not read as one company's bad week. Plasma's own access vault, in June 2025, drew a different complaint: not an insider pre-fill but raw whale and bot concentration, with the cap raised from $250 million to $500 million and then to $1 billion, the top ten wallets taking a large share of the first round, and gas-fee bidding wars just to land a transaction. Berachain's Boyco, built on Royco, drew a third: it peaked at roughly $3.1 billion in pre-deposits across more than 150,000 wallets, then shed most of that the moment the mainnet launched and the rewards stopped, because the capital was there for the incentive rather than the chain. The protocols and the failures are different, but the root is shared. The terms that decide what a depositor actually gets, when the cap fills, who filled it, whether it can be raised, whether the money is locked, were resolved somewhere the depositor could not see until the capital was already in.

Underneath those episodes is the same short list of questions. Could the cap be filled, or was it filled, before you could see the window. Is there a per-wallet limit, or is it a latency race a whale wins. Can the stated cap be raised after you deposit, changing the share you expected. Can you see the actual reserves while the window is open, or only a headline number. Can you get your money back at par when you want it, or is it locked past the token event. Each of those has an answer in advance, if the protocol lets you check it.

The four things a depositor can actually check

None of this is a reason to write off pre-deposits. The better response is a short checklist you can run on any window, ours included, before committing a dollar.

Where our Genesis window sits, worst facts first

We are opening a capped Genesis pre-deposit window around the publication of our first external audit, so we are an interested party here, and we will run our own window through the same four checks, including the parts that do not flatter us.

Start with the fact that makes the rest checkable. Our whole book is small and fully public. As of this writing the protocol holds about $1,114 of reserves across its Peg Stability Module contracts, backing about the same amount of kUSD, spread across roughly seven holders. Every dollar of that is enumerated in an hourly proof of reserves signed with an EIP-191 key. The on-chain leg is readable straight from Base at /api/por, and the signature and its freshness are checkable at /verify, with no attestor in between. A hidden pre-fill is not possible against a reserve book that is fully on-chain and this small: you can read every address before you commit anything. The same book, and the gap between what we advertise and what holders actually earned, sits first and currently worst on our honesty index. We would rather show a small number that is real than a large one we could not stand behind.

On entry, there is no cap race today. Right now the founding multiplier accrues to kUSD minted one to one against USDC at the live PSM, and it steps down on a published weekly schedule anyone can read on the Opal ladder, not by who lands the first transaction. The rule those points settle into is published in advance at /opal/anchor-tier: a fixed pool of 50,000,000 KERNE, 5 percent of supply, that a higher multiplier can redistribute but never enlarge, converting only if and when a token event occurs, which is not scheduled or guaranteed. On the way out, kUSD redeems at par through the same PSM with no lockup and no fixed exit window. We will not overstate that: the PSM, like any responsible peg module, keeps an emergency pause held by the multisig, so the accurate claim is par redemption with no lockup, not a promise that a pause can never happen. The dedicated Genesis escrow we designed to be non-pausable is not deployed yet and does not go live until it is built and inside the audit scope. We are not going to describe a future contract as if it were a current one.

The boundaries belong in the same breath, because a pre-deposit piece that hid its author's own gaps would be the failure it describes. We are pre-audit. Hexens delivered the initial report of our first external review on July 20, 2026, remediation is underway, and the final report is pending, so the code is not yet through a completed audit, and we say so on every surface. The delta-neutral hedge runs at pilot scale against a small, disclosed, founder-custodied float, and that leg is self-reported and signature-bound rather than independently re-derivable on-chain; an independent attestation of it is being scoped. kUSD is a synthetic dollar, not a payment stablecoin, and it pays no yield for being held; yield exists only through the separate skUSD staking receipt, which is gated on the audit. The skUSD APY we model is 13.1 percent at target leverage, against a published through-cycle band of 8 to 9.4 percent and a realized figure near 0.19 percent from a single test distribution so far, and all three of those numbers sit together on the index rather than the first one standing alone. The specific cap and seat count of the founding window are set in the anchor deposit letter and remain subject to final terms, so we are not printing those numbers while they can still move. What is fixed and published is the points rule and the pool.

None of that makes our window the right thing for any particular person to deposit into, and it is not an offer. It makes the window checkable, which is the only claim we are making against the pattern above. Stable's timeline did not matter because the vault filled quickly. It mattered because by the time the public could look, the answers to those four questions were already set and unfavorable, and an Etherscan post-mortem was the only tool left. A window you can read in full, priced on a published schedule, redeemable at par, with a live proof of reserves while it is open, gives you those answers before you decide instead of after, and you can check every one of them without our help.

Related reading: the mirror-image live case where a real 24 percent supply drop turned out to be a redemption honored at par, our USD.AI teardown, and this week's three phantom nine-figure TVL crashes that were dashboard artifacts rather than outflows, TVL is not a proof of reserves. The window itself is at /genesis.

Sources, all as of July 22, 2026. Stable Phase 1: The Block, "Stable's Phase 1 pre-deposit hits $825 million cap" (Oct 24, 2025), and The Defiant, "Stable controversially pre-fills $825 million vault" (Oct 2025), for the $825M cap, the roughly 22-minute fill, the pre-announcement deposit timestamps, the roughly $500M across about ten wallets described as linked to the vault owner, the single-wallet concentration, the sub-300 participant count, the absence of per-wallet caps, and Stable not responding to a request for comment; the Veda contract-hiding quote is from the same Defiant reporting. Stable Phase 2 (KYC, per-wallet caps, an announced start, and the Hourglass front-end crash, denial-of-service, deposit pause, and direct-deposit refunds) from The Block's Phase 2 coverage and The Defiant's pause report. Plasma's cap raises and concentration from The Defiant, and the minimum-40-day post-window lock from Plasma's own public-sale FAQ; Berachain Boyco's peak and post-launch decline from Sentora's case study. Kerne figures resolve to live endpoints and are point-in-time: reserves and holder count from the hourly signed proof of reserves at /api/por/signed and its on-chain leg at /api/por; the modeled APY, the 8 to 9.4 percent through-cycle band, and the realized figure from /api/apy, the honesty index, and our published feasibility band; the points floor and fixed pool from /opal/anchor-tier; audit status from /security/audits. This is not investment, financial, or legal advice, and it is not a recommendation to deposit into any window, including ours. Nothing here is an offer of any token or a solicitation to buy or sell any security. Recompute before relying on any figure here, including ours.