Text size
Security

Why Crypto Bridges Keep Getting Hacked — and What an "Audited" Badge Doesn't Tell You

The Cysic bridge exploit and Hacken's Q3 report ($1.18B stolen) show where crypto security breaks. A beginner's guide to bridge risk and reading audits.

Not financial advice. This article is for general information and education only. It is not a recommendation to buy, sell or hold any asset. Crypto assets are volatile and you can lose money; do your own research.

On Saturday, October 17, an attacker minted 10 million unauthorized CYS tokens through the cross-chain bridge run by Cysic, a zero-knowledge proof network, sold most of them and moved the proceeds to Ethereum within about 35 minutes. The same weekend, blockchain security firm Hacken published a report putting third-quarter 2026 crypto hack losses at $1.18 billion — and that most of the losses came from systems that security audits never looked at. This article uses both stories to explain why bridges are such a frequent target and how to judge security claims before you trust a protocol with your money.

What happened at Cysic

According to The Block's on-chain reconstruction and Cointelegraph's report:

  • 8:31 a.m. ET: using Cysic's bridge, the attacker created 10 million new CYS tokens on BNB Smart Chain.
  • Within about 30 minutes of the mint, the receiving address sold 7.2 million of the tokens, mostly through a DEX aggregator, for roughly 588,000 USDT and about 220 BNB.
  • The proceeds were bridged to Ethereum through three other cross-chain services, arriving as about 306.6 ETH — roughly $727,000 at Sunday's prices.
  • At 9:47 a.m. ET Cysic added the attacker's address to a blacklist in the CYS token contract, locking the 2.8 million tokens still there — but that came 76 minutes after the exploit started, and the ether had already left.

The Cysic Foundation said the attacker "gained unauthorized access to a database" that supported the bridge's services, and then abused a weakness in how the bridge processed messages to push through a fake mint instruction. According to the foundation, the damage stayed within the bridge's message handling and minting, the Cysic mainnet was untouched, and no existing user funds were taken. The bridge remains closed pending a security review.

How a bridge works — and why it's fragile

Blockchains cannot natively read each other's data. A bridge solves this by locking tokens on one chain and issuing an equivalent on another, or by burning on one side and minting on the other. Somewhere in the middle, a piece of software — often run by a small set of servers, validators or signers — has to decide: "Yes, this deposit really happened, so mint the tokens over there."

That decision point is the weak spot. Whoever can convince the bridge that a deposit happened can create tokens out of thin air. Attackers have achieved this in different ways over the years:

  • Compromised signer keys, where the attacker gains control of enough of the keys that approve cross-chain messages.
  • Bugs in verification logic, where the smart contract accepts a message it should have rejected.
  • Off-chain infrastructure, such as the database and processing services Cysic described, which sit outside the smart contracts entirely.

Because a bridge often holds large pools of locked assets, or can mint tokens with real market value, a single flaw can be extremely costly. Our DeFi guide already flags bridges as "historically major hack targets"; the Cysic incident is the latest example.

Hacken's Q3 report: the problem is scope

CoinDesk's write-up of Hacken's report contains the more important lesson. Hacken counted 107 incidents and $1.18 billion in losses in Q3, up 55% from the previous quarter. Of that, $906.8 million — about 77% — stemmed from three causes:

  1. Compromised suppliers or developer tools (supply-chain attacks).
  2. Publicly known security fixes that were never installed.
  3. Parts of the system the latest security review did not cover.

The biggest loss of the quarter came in September at the exchange Bitget: Hacken put it at $352 million or more, with attackers getting into withdrawal systems through a bug in security software Bitget bought from an outside vendor. Hacken also counted more than $35 million lost by three bridges and cross-chain protocols in a six-hour July exploit.

Hacken's co-founder Andrii Matiukhin summarized it bluntly to CoinDesk: "The problem is always scope." Audits are expensive, so teams often pay to review only the most critical smart contracts. Wallets, signing systems, servers and third-party software can be left out — and those are exactly what attackers target.

What an audit badge does and doesn't mean

An audit is a time-limited review of specific code by a specific firm. It is useful, but it is not a guarantee. When a project advertises "audited", check:

  • What was in scope? Good audit reports list exactly which contracts or components were reviewed. If the bridge relayers, admin keys or front end are not listed, they were not checked.
  • When was it done? Code changes after an audit are not covered. Compare the audit date with the project's latest upgrades.
  • Were the findings fixed? Reports usually list issues and whether the team resolved, acknowledged or ignored them. Hacken noted that critical issues are sometimes downgraded and left unfixed.
  • Who controls the admin keys? Hacken found more than $700 million at two major venues depended on a three-of-four signing arrangement or on administrative keys held by unidentified people. Small multisigs and anonymous key holders concentrate risk.
  • Is there more than one audit? Multiple independent reviews, a public bug bounty and a track record of handling incidents are stronger signals than a single badge.

Practical steps for users

You cannot audit a bridge yourself, but you can limit how much a failure could cost you:

  • Minimize time in transit. Bridge only what you need, and do not leave large balances in "wrapped" tokens that depend on a bridge's solvency.
  • Prefer native assets and issuer-run transfers where available. For example, some stablecoin issuers provide their own cross-chain transfer services, which removes a third-party bridge from the path.
  • Watch for pauses and blacklists. As Cysic showed, a project can freeze tokens on one chain, but it cannot claw back assets once they have been converted and moved elsewhere.
  • Beware of fake "refund" or "recovery" sites after a hack. Scammers routinely impersonate exploited projects. Revisit our security essentials for how to spot them.
  • Size positions to the risk. A newer bridge with a small team and a single audit deserves a smaller allocation than infrastructure with years of incident-free operation.

The bottom line

The Cysic exploit fits a pattern: the attacker did not need to break any blockchain, only to persuade one piece of bridge infrastructure that a fake message was real. Hacken's data shows the same pattern at industry scale — most of the $1.18 billion lost in Q3 came from suppliers, unpatched bugs and systems left out of audits. Treat "audited" as the start of your questions, not the end of them.

Sources: The Block, Oct. 18, 2026; Cointelegraph, Oct. 18, 2026; CoinDesk, Oct. 18, 2026.

#bridges#hacks#defi#audits#security#cross-chain

Spotted an error? Tell us and we will correct it. See also our disclaimer and beginner guides in Learn.

Related articles

← All articles