A zero-knowledge proving network has paused its cross-chain bridge after an attacker walked through the bridge’s own withdrawal path and minted 10 million CYS tokens on BNB Chain. The incident is a compact lesson in what a block explorer can and cannot settle, and in why the real damage is often measured in minutes rather than tokens.
What the chain actually shows
Two transactions bracket the event, and both are public. At 12:31:38 UTC on 17 October, a transaction calling the bridge’s withdraw function minted 10,000,000 CYS from the null address to a single receiving account. A mint from the null address has no prior holder, which is what makes it checkable: the tokens did not leave a treasury or a reserve. Roughly 76 minutes later, at 13:47:52 UTC, a separate multisig transaction executed, matching the timing Cysic gave for blacklisting the receiving address. The mint amount and both timestamps are facts anyone can verify directly on a block explorer.
The gap is the point
By the time the blacklist landed, Cysic said only about 2.8 million of the 10 million minted tokens were still sitting at the first address. The other 7.2 million, about 72 percent, had already moved. Ten million CYS equals one percent of the one-billion maximum supply and a larger slice of the supply actually in circulation, so this was not a rounding error. The distinction matters because a fixed token count dilutes holders more when the actively traded float is smaller. The important number, though, is not the mint; it is the window. A bridge mint can in principle be contained with a blacklist, but only for the balance that is still in the flagged address when the blacklist fires.
What the company says, and what it does not
Cysic’s preliminary account attributes the mint to unauthorized access to a database that supports the bridge, combined with a flaw in the message-processing logic that allowed a forged mint message to pass. That is a company statement about root cause, not a finished investigation: the full attack path is still under review, and a security firm has been engaged to trace the moved funds. The result is that the two verifiable facts — the mint and the blacklist — sit on-chain, while the explanation for how the mint happened rests, for now, on the foundation’s word. Cysic was also explicit about scope. The incident was confined to the bridge’s message-processing and minting functions; existing user assets were not taken, and the mainnet, including consensus, block production, and EVM execution, was unaffected. The bridge stays closed until a full security review is complete, with no reopening timeline offered.
Why bridges keep showing up in these stories
Bridges remain a concentrated attack surface because they must reconcile two chains with different security assumptions, and a single message that convinces the destination chain to mint is enough to do damage. The same pattern appeared in Fetch.ai’s SingularityNET bridge exploit in September. For a holder, the takeaway is specific: a mint that happens is not the same as a mint that gets contained. The loss is set by what left the flagged address before anyone could respond.
Source: BlockchainReporter.