The perpetual contract for SK Hynix shares (ticker xyz:SKHYNIX), traded on the Hyperliquid platform, lost 17.9% of its value in a short period of time, before recovering just as rapidly to previous levels. This incident revealed important nuances in the operation of decentralized infrastructure solutions and their dependence on external data.
The cause of such a sharp movement was an anomalous trade during the pre-market session of the South Korean trading platform NXT. Just one share of SK Hynix was sold at a price of 1.272 million won, triggering a movement of about 30% and leading to a temporary halt in trading of this asset on the traditional exchange. It was this single, essentially erroneous order that became the trigger for a cascade of events in the cryptocurrency ecosystem.
Price Transmission Mechanism: How a Single Error Became an Avalanche
The contract on Hyperliquid does not have its own market depth comparable to traditional exchanges. Its price is pegged to external sources via an oracle. In this case, the oracle transmitted the price from NXT to the blockchain, which recorded the anomaly. Following this, prices also began to fall on Binance due to the actions of arbitrage bots seeking to profit from the difference in quotes between platforms. The mechanism worked like a chain reaction: a single error in a low-liquidity market caused a domino effect across the entire ecosystem.
It is important to emphasize that the xyz:SKHYNIX contract itself is not a product of the Hyperliquid team. It was launched and is maintained by a third-party team, XYZ, under the HIP-3 standard. This model allows any developer to create their own market on Hyperliquid but places responsibility on them for configuring price parameters, including the selection of data sources for the oracle.
The Key Role of the Mark Price
The central element in this story is the mark price—the calculated price by which the platform computes profits, losses, and forced liquidations of positions. According to the HIP-3 specification, the deployer (market creator) can influence this price by providing two of the three values used to calculate the median. The third value is generated by the protocol itself based on data from the last trade and the best bid and ask orders.
Thus, if the protocol shows one value and the deployer provides two others, the final mark price can differ significantly from the real market price. This is exactly what happened in this case: the market operator, relying on oracle data, transmitted an anomalous price, which led to the contract's 17.9% crash. This incident clearly demonstrates that when dealing with third-party markets on Hyperliquid, any complaints about sharp movements should be directed not at the platform itself, but at the team managing the specific contract and its price settings.
Expert opinion: This case is a stark reminder of the risks associated with using oracles and dependence on external, sometimes low-liquidity markets. For traders, it is a signal to carefully study the architecture of the contracts they trade and understand who sets the mark price and how. For platforms like Hyperliquid, it is a challenge—how to maintain decentralization and openness while simultaneously protecting users from such anomalies.