Ethereum block builders captured roughly $1.78 million of the $3.42 million drained from K3 Capital's sBOLD vault on Nov. 18, 2026, more than the original attacker walked away with, according to onchain data cited by CoinDesk and traced by Bitquery. The eight-minute drain turned into a bidding war in which a copycat trading bot paid most of its proceeds for the right to land transactions first.
Primary Sources & Key Developments
Bitquery's transaction review found the vault was emptied between 12:25:23 and 12:33:35 UTC, a window of about eight minutes. The firm traced 97 transactions sent from six wallets across 37 blocks. Blockaid's initial alert put the loss at roughly $3.43 million across about 34 transactions, a narrower count that Bitquery's later reconstruction expanded.
The mechanics matter more than the totals. An automated trading bot copied the attack and front-ran the flow, extracting $1.91 million, or 56% of the stolen funds, and sending 57 of the 97 drain transactions. To get those transactions included, the bot bid aggressively for block space. Its payment to Ethereum block builders climbed from 12.5 ETH on its first drain to 44 ETH for one block near the end of the sequence.
Titan built most of those blocks, according to Bitquery, with BuilderNet, Ultrasound and Quasar building the rest. The original attacker wallet 0xdc0e and a wallet it funded ended up with about $520,000 after the bidding contest, a CoinDesk review of the transactions found.
Why Did Ethereum Block Builders Earn More Than the Attacker?
Because the race was decided in the mempool, not inside the vault. Once the first drain transaction was public, any searcher could replicate the call, and the only scarce resource left was ordering. Builders sell ordering, so value that would normally accrue to an exploiter was redirected into priority fees and builder payments within seconds.
Bitquery said it found no loan liquidations in its transaction review, despite initial alerts pointing to self-liquidations. That removes one common explanation for the size of the fee spend and leaves competitive bidding as the main driver.
By the Numbers: Where the $3.42M Went
The distribution of the $3.42 million is unusual for a DeFi exploit. In many incidents the attacker keeps the bulk of the proceeds. Here the infrastructure layer, meaning block builders, took the largest single share.
Key Benchmark Comparison (Party | Reported amount | Share of $3.42M drained)
- Ethereum block builders: Reported amount: ~$1.78M | Share of $3.42M drained: ~52%
- Copycat bot (gross extraction): Reported amount: $1.91M | Share of $3.42M drained: 56%
- Original attacker wallet 0xdc0e and funded wallet: Reported amount: ~$520,000 | Share of $3.42M drained: ~15%
- Total drained from sBOLD vault: Reported amount: $3.42M | Share of $3.42M drained: 100%
Categories overlap: the bot's gross extraction was largely paid out again as builder payments, which is why the rows sum above the total. Bitquery framed its investigation as the bot and builders keeping 59% of the $3.42 million. The vault held nothing else afterward, the firm said.
The fee escalation is the clearest signal of competition. A 12.5 ETH opening bid rising to 44 ETH for a single block implies the bot expected the marginal block to be worth more than the marginal bribe, a calculation that only works when the underlying transaction is a one-time, non-repeatable extraction.
Audit Trail & Counterparty Implications
K3 Capital paused the sBOLD contracts after the drain. The audit trail now becomes the central question. ChainSecurity reviewed sBOLD in May 2025 and identified an accounting issue where the vault ignored collateral below a set threshold, which its example put at 7,500 BOLD. The report marked the issue as fixed in a later version of the code.
It remains unclear whether the Nov. 18 attack exploited that previously identified accounting vulnerability, whether the reviewed fixes were deployed to the live vault, or whether a different accounting weakness was involved. Until K3 Capital or ChainSecurity confirms which code was running, the incident sits between a known-and-patched finding and an unknown one.
For counterparties, the builder concentration is the practical takeaway. A handful of entities, Titan most prominently in this sequence, controlled inclusion for the blocks that mattered. That gives builders visibility into, and pricing power over, exactly the kind of urgent flow that exploits generate. Vault operators and their auditors now have to model builder fees as a first-order cost of an incident, not an afterthought.
The market backdrop is not helping. The Crypto Fear & Greed Index read 10 out of 100 on Nov. 19, 2026, deep in Extreme Fear, a reading that signals broad risk aversion across crypto markets.
What Do Analysts Expect Next?
Expect three lines of inquiry. First, a public post-mortem from K3 Capital reconciling the 34-transaction Blockaid count with Bitquery's 97, since the gap implies either repeated calls or internal transfers that the first alert missed. Second, disclosure on whether the May 2025 ChainSecurity fix was live, which determines whether this is a regression or a new bug class.
Third, and least discussed, pressure on builder transparency. Bitquery could attribute blocks to Titan, BuilderNet, Ultrasound and Quasar, but the payment flow itself is only visible because the bot overpaid in a way that stood out. If 52% of a hack's value can route to block space in eight minutes, the economics of Ethereum MEV infrastructure are doing exactly what they were designed to do, and that is the part no audit fixes.
Frequently Asked Questions
How much did Ethereum block builders receive from the sBOLD vault hack?
Block builders received approximately $1.78 million, or about 52% of the $3.42 million drained from K3 Capital's sBOLD vault on Nov. 18, 2026, according to onchain data cited by CoinDesk and traced by Bitquery.
Who ended up with the stolen sBOLD funds?
A copycat automated trading bot extracted $1.91 million, or 56% of the stolen funds, but paid most of it back out as builder payments. The original attacker wallet 0xdc0e and a wallet it funded retained about $520,000, while block builders took roughly $1.78 million.
Was the sBOLD vault hack caused by a known ChainSecurity audit finding?
That is unresolved. ChainSecurity audited sBOLD in May 2025 and flagged an accounting issue where the vault ignored collateral below a set threshold, marking it fixed in a later code version. It remains unclear whether the Nov. 18 attack used that vulnerability, whether the fixes were deployed, or whether a different accounting weakness was involved.
Related Research
Primary Sources
- Ethereum block builders received the most in a $3.4 million vault hack, data shows (CoinDesk)
- Crypto Fear & Greed Index — Alternative.me (Alternative.me)
- K3 Capital sBOLD Hack: Bot and Builders Kept 59% of $3.42M - Bitquery (Verified Web Source)
Disclosure: This market analysis is provided for informational and educational purposes only and does not constitute financial or investment advice.