Native Core CLOB
Summary
Ketju does not approve Native Core CLOB under its class rule for protocols on unapproved chains: every Pool balance sits on Native Core, a chain Native runs that Ketju has not reviewed. Native Core is an order-book chain that Native runs for its own matching engine, and Native Pool is the earn product on it. A depositor moves WBTC, USDC, ETH, USDT, BNB, or QQQB from an EVM chain into a Native Core balance and places it in the Pool. In return the depositor receives a share of the trading and funding fees that verified market makers and Native Relay earn with that inventory. DefiLlama showed six Pool balances worth about $2.64M on 2026-10-16. DefiLlama listed the protocol on 2026-08-31 under the Native parent, with no TVL series of its own. We made no finding against Native. Ketju classes the exposure as a liquidity pool because the deposit is market-making inventory; it is not a loan at a posted rate.
- Ketju approves Native Core after a chain review with no open review trigger, which requires Native to publish a verifiable validator set and withdrawal signer quorum
- Native launches a verifiable Pool deployment on an approved chain and keeps it above the size limit for 30 days
Ketju checks this memo every night and opens a review if its venues or files show a warning or a cited document changes. The first confirmation is due 2027-01-16.
Research
How it works
Native Core is a dedicated app chain, chain id 696969, that stores the full order book, matches orders first in first out, and claims 50 millisecond block finality. Assets do not originate there. A depositor calls deposit() on a DepositWithdrawVault contract on Ethereum, BNB Chain, Base, or Arbitrum and pays a one-time account activation fee worth about $1. Native credits a Native Core balance a few minutes later, and Native Pool takes that balance into an earn balance. Demand for capital on the chain funds the yield. Verified market makers draw on the inventory to quote and settle on the order book and pay funding fees. Native Relay uses the inventory to fill swaps on EVM chains and pays trading fees.
Native distributes yield; balances do not accrue it. At 00:00 UTC each day the Pool snapshots every earn balance. Native’s operations team then decides the amount to distribute per asset, and each account gets its pro rata share. The Pool has no interest-rate setting. The projected rate DefiLlama reports is the last distribution annualised over the gap since the one before. On 2026-10-16 DefiLlama showed WBTC at about $1.43M, USDC $0.70M, ETH $0.31M, USDT $0.16M, BNB $0.03M, and QQQB $0.02M, with projected rates from 0.1% to 5.8%. The depositor keeps the asset’s own price risk. The depositor also takes on the credit accounts the protocol extends to market makers. The Risks page says liquidations can fail in fast markets and leave bad debt, which is why Ketju classes the exposure as a liquidity pool and not as stable lending.
Who controls it
The docs describe consensus and finality on Native Core. They do not name who runs the validators or how many there are, and Native runs the explorer at app.native.org itself. The protocol provisions credit accounts; market makers cannot open them on their own. An operator can freeze a credit account, and a frozen account may only cancel orders. An operator releases withdrawals from Native Core back to an EVM chain with an operator-only call on the vault. That call carries an M-of-N signature set from what the docs call Native’s signer network. The docs do not publish M, N, or the signers. The vault admin can pause deposits per token and for a whole vault. The operations team controls a withdrawal_paused flag in the Pool config. Native marks the published contract source as unlicensed and offers it for review only.
Losses and incidents
The audits page lists seven reports from February 2023 to October 2025 on Native’s EVM contracts and its V2 core contract. The latest are by Sherlock in June 2025 and WatchPug in October 2025. None names a review of the chain’s consensus, the signer network, or the Pool’s accounting. The docs mention a bug bounty but give no terms. We found no exploit or loss on Native Core. The Pool replaced earlier Fixed-Term and Flexible Earn products, and the pools first appeared on DefiLlama’s yield list in October 2026, so the operating history is short. Because the class rule applies, we did not verify the validator set, the signer quorum, or how Native underwrites the credit lines.
Withdrawals
A withdrawal takes two steps. First the depositor withdraws from the Pool to a Native Core balance. A scheduled withdrawal is free but waits withdraw_pending_seconds, which was 259,200 seconds, or 3 days, for every asset on 2026-10-16. It earns nothing while pending, and the depositor can cancel it only during the wait. An instant withdrawal pays instant_fee_bps, 100 basis points or 1% on 2026-10-16, subject to availability. Only one withdrawal per asset can be pending at a time. Second, the depositor signs a withdraw action to Native Core, which debits the balance at once. The depositor then waits for Native to release tokens from the destination chain’s vault. Each chain’s vault releases only what it holds. The docs’ own example shows 664,110 USDT of headline capacity, of which 1 USDT sat on Base. Neither step can be reversed once it lands. Native holds a deposit that fails validation for manual review and does not return it automatically. The user docs say withdrawals settle instantly or within T+1, depending on amount and liquidity. Native’s chain, Native’s operators, and Native’s signers handle every step.
Class rule
Ketju applies the class rule for protocols on unapproved chains to Native Core CLOB. The rule covers protocols whose live yield venues settle only on a chain that Ketju has not approved after a chain review. Native Core is the first case: an app chain with no Ketju review and no public validator set to review. Every balance change, distribution, and withdrawal depends on that chain’s consensus and on the bridge that moves assets in and out. Here one company runs the chain, the bridge signers, and the operator. On an approved chain the same inventory could sit in a reviewed pool, but no Native Pool deployment exists that an advisor can reach without Native Core. Ketju will review the memo again only if it approves Native Core after a chain review, or if Native launches a material Pool deployment on an approved chain. Either event would start a full review; neither would make the Pool approved. The pools are also below the size limit for a full review.
Sources
Each claim above comes from one of these sources. Where Ketju could not verify a number, the memo states that.
- Native docs: what is Native · primary · accessed 2026-10-16
Supports: sovereign on-chain matching engine, Native Core, Pool, Pro, Relay modules - Native docs: Native Core · primary · accessed 2026-10-16
Supports: FIFO on-chain CLOB, 50ms block finality, Native explorer - Native docs: Native Pool · primary · accessed 2026-10-16
Supports: earn product, yield from market maker utilisation and Relay fees, scheduled and instant withdrawal - Native docs: Native Pool yield · primary · accessed 2026-10-16
Supports: verified market makers pay funding fees, Relay trading fees, one scheduled withdrawal pending at a time - Native docs: Native Pool FAQ · primary · accessed 2026-10-16
Supports: yield variable, Fixed-Term and Flexible Earn replaced, ability to pause product activity - Native docs: CLOB · primary · accessed 2026-10-16
Supports: dedicated app chain, price-time priority, full order book on chain - Native docs: on-chain verifiability · primary · accessed 2026-10-16
Supports: consensus, deterministic matching, finality - Native docs: risks · primary · accessed 2026-10-16
Supports: bad debt risk, liquidations may fail, bug bounty mentioned - Native docs: account types · primary · accessed 2026-10-16
Supports: credit accounts provisioned by the protocol, USD credit line, operator unfreezes - Native docs: deposit and withdraw · primary · accessed 2026-10-16
Supports: funding spans two chains, no reversal, failed deposits held for manual review - Native docs: withdraw · primary · accessed 2026-10-16
Supports: per-chain vault capacity, balance debited before release, chain id 696969 - Native docs: vault contract · primary · accessed 2026-10-16
Supports: vault addresses on four chains, operator-only withdraw with M-of-N signer network, deposit pause flags - Native docs: Native Core Pool API · primary · accessed 2026-10-16
Supports: withdrawal_paused flag, withdraw_pending_seconds, instant_fee_bps, daily snapshot time - Native docs: Native Core Pool yield · primary · accessed 2026-10-16
Supports: yield distributed not accrued, operations decides the amount, no interest-rate setting, projected rate definition - Native docs: Native Core Pool withdraw · primary · accessed 2026-10-16
Supports: scheduled versus instant, cancel only during the wait, one withdrawal per asset in flight - Native docs: audits · primary · accessed 2026-10-16
Supports: seven audits 2023 to 2025, Sherlock June 2025, WatchPug October 2025 - DefiLlama: Native Core CLOB protocol page · secondary · accessed 2026-10-16
Supports: chain, category, listing date, parent protocol, no TVL series
The asset and the chain
This memo covers the protocol. A position also holds an asset and settles on a chain. Whichever of the three holds the most power over the position decides who can freeze it. These tables describe control and do not judge quality.
| Chain | Verdict | Freeze | What it can do |
|---|
| Asset | Freeze | Who can freeze it |
|---|---|---|
| WBTC | Issuer can freeze | WBTC tracks Bitcoin, but it is a custodial claim: BitGo holds the underlying BTC and can be compelled to act on it. Unlike BTC, it depends on an issuer. |
| USDC | Issuer can freeze | Circle issues USDC and backs it with bank deposits and T-bills. Circle can freeze addresses, and does so when law enforcement asks. |
| ETH | No freeze key | ETH has no issuer, and no party can freeze or seize it on chain. |
| USDT | Issuer can freeze | Tether issues USDT and has frozen addresses on request. Its reserves are attested less transparently than USDC’s. |