JarValley

Market Prices

BTC Bitcoin
$79,760 -1.34%
ETH Ethereum
$2,458.55 -1.43%
SOL Solana
$101.93 -2.21%
BNB BNB Chain
$720.1 -0.12%
XRP XRP Ledger
$1.41 -3.65%
DOGE Dogecoin
$0.0848 -5.39%
ADA Cardano
$0.2146 -3.33%
AVAX Avalanche
$7.39 -1.78%
DOT Polkadot
$0.8586 -3.23%
LINK Chainlink
$11.71 +0.01%

Event Calendar

{{ๅนดไปฝ}}
28
03
unlock Arbitrum Token Unlock

92 million ARB released

30
04
upgrade Celestia Mainnet Upgrade

Improves data availability sampling efficiency

18
03
unlock Sui Token Unlock

Team and early investor shares released

12
05
halving BCH Halving

Block reward halving event

22
03
unlock Optimism Unlock

Circulating supply increases by about 2%

10
05
upgrade Ethereum Pectra Upgrade

Raises validator limit and account abstraction

15
04
halving Bitcoin Halving

Block reward reduced to 3.125 BTC

08
04
upgrade Solana Firedancer

Independent validator client goes live on mainnet

Tools

All โ†’

Altseason Index

41

Bitcoin Season

BTC Dominance Altseason

Market Cap

All โ†’
# Coin Price
1
Bitcoin BTC
$79,760
1
Ethereum ETH
$2,458.55
1
Solana SOL
$101.93
1
BNB Chain BNB
$720.1
1
XRP Ledger XRP
$1.41
1
Dogecoin DOGE
$0.0848
1
Cardano ADA
$0.2146
1
Avalanche AVAX
$7.39
1
Polkadot DOT
$0.8586
1
Chainlink LINK
$11.71

๐Ÿ‹ Whale Tracker

๐Ÿ”ต
0xe320...2fd6
12h ago
Stake
2,709,524 USDC
๐Ÿ”ต
0x6783...3b7c
1d ago
Stake
4,145,253 USDT
๐Ÿ”ต
0x16bb...8ba5
12m ago
Stake
2,870,866 USDC
Gaming

Solana Neobank Avici Drains $650K: The Self-Custody Contract Was the Attack Surface

0xHasu

The numbers hit first. $650,000 drained from user custody vaults. AVICI token down 40% to $0.24. Over 9,000 users exposed. The attack was still running when tracker updates landed โ€” two hours in, balances kept climbing. Avici's response: "We are aware and monitoring." No contract pause. No compensation plan. No confirmation of the attack vector.

The mechanism deserves sharper focus than the dollar figure. Attackers used "carefully crafted signature packages" to grant themselves administrator status over custody contracts. To be precise: the deployed code contained a privileged path contradicting Avici's own documentation. The docs promised only user wallets could move funds. The bytecode disagreed.

This gap between narrative and implementation is where the money went. It is why this incident fragments into something larger: a Solana neobank, a self-custody promise, and a permissions failure.

Avici occupies a fragile niche โ€” the crypto neobank. Users lock USDC into custody contracts, receive a Visa card issued through Third National, and spend against their own collateral. Self-custody is the marketing anchor. Your assets stay in your control while you swipe at merchants.

The architecture creates layered dependencies. Solana for settlement. USDC for the float. Third National for card issuance. The smart contract as the custody layer. That last piece failed.

This is not isolated. Midnight, a comparable project, suffered a bridge vulnerability in July and its token hit historical lows. BONK DAO weathered a governance attack earlier this year. The pattern is systematic. Solana's banking sector is under recurring pressure.

Now the technical read, based on available evidence.

The exploit signature: crafted signed transactions that elevated attacker-controlled addresses to admin status over per-user vaults. Not a flash loan trick. Not oracle manipulation. A permissions failure. Either the attacker invoked an admin function without authorization, or obtained the administrator keys themselves. Both scenarios damage trust.

From a design perspective, the defect is verifiable. Documentation promised a single withdrawal path. The contract exposed additional privilege functions. My audit background has a term for this โ€” the narrative-reality gap. Auditors hunt for it. Attackers hunt for it. This time the attacker found it first.

The security model failed at the assumption level, not just the implementation level.

Economic math sharpens the picture. $653,000 stolen against a market capitalization of roughly $3.26 million. The stolen sum equals about 20% of total token value. A protocol can absorb a 2% loss. It cannot absorb 20%.

The token collapse to $0.24 reflects a correct market read: AVICI's value is trust. The governance token carries no independent economic weight. No fee burns. No staking requirements. No disclosed revenue-sharing. Value derived from a belief that the treasury is solvent and user funds are safe. That belief is now broken.

The withdrawal pattern is revealing. The attacker drained accounts individually rather than clearing a shared pool. That points to automation. Scripted extraction. Methodical. This suggests a sophisticated operator who mapped the account structure in advance โ€” or access to internal logic, meaning a compromised admin key rather than an unpatched exploit. Inference, not confirmation. But the distinction matters for users deciding whether to trust remaining funds.

Liquidity risk compounds the loss. A token this size runs on thin order books. If market makers withdraw, price discovery deteriorates. Panic selling into illiquidity creates a loop: lower prices trigger more panic, more panic dries up liquidity. The 40% drop is not necessarily the bottom.

The underreported angle is not the $650K. It is the damage to the self-custody narrative.

Most retail users cannot distinguish an admin key from a user withdrawal function. They read documentation. Avici's documentation said user-only withdrawals. The deployed reality was different. This breach did not simply steal funds. It demonstrated that in the current generation of neobank contracts, self-custody is a marketing label, not a technical guarantee.

The ripple effect is measurable. Competing protocols now carry a verification burden when advertising self-custody. The burden is not optional. Regulators are watching. When documentation says "only your wallet can move funds" and that statement proves false, deceptive advertising questions follow.

Third National, the card issuer, holds the next decision. If the issuer walks away, Avici loses the ability to operate its product. One public statement from the issuer determines the project's survival faster than any on-chain recovery.

Security congestion is spreading across this niche. Every "self-custody" claim just got more expensive to verify.

Watch three signals. First: is the attacker wallet still active? If the drain paused, losses cap near $650K. If it continues, assume the contract remains compromised. Second: does Third National issue a statement? Silence signals a strained relationship. Third: monitor Solana DeFi TVL over the next two weeks. If capital exits the ecosystem's custody venues, this attack becomes a liquidity event, not just a security story.

The open question is structural. After a custody breach disguised as self-custody, what does the term actually communicate? The next protocol to use that word must prove it. Proof happens in code, not in documentation. It should have been that way from the start.

Fear & Greed

74

Greed

Market Sentiment

Gas Tracker

Ethereum 28 Gwei
BNB Chain 3 Gwei
Polygon 42 Gwei
Arbitrum 0.5 Gwei
Optimism 0.3 Gwei

๐Ÿ’ก Smart Money

0x7dc3...278c
Market Maker
+$0.9M
72%
0x04c4...86c8
Experienced On-chain Trader
+$4.1M
80%
0xb775...b79f
Market Maker
+$3.3M
84%