Lighter's Nov. 16 security upgrade described how exposed trading keys enabled an account drain through public-pool shares, putting a potential Hyperliquid exploit in perspective.
Lighter said that withdrawals had resumed after its safeguard allowed about 5 million USDC to be returned.
Hyperliquid has roughly $5.7 billion in open interest, representing about 55% of the open interest in the DefiLlama ranking as of Nov. 16.
Recent findings suggest that depending on the account permissions, a compromised agent could cause losses through unauthorized orders and liquidation in the accounts it can trade. Wider losses would depend on the affected positions, liquidity providers, and market depth.
Lighter said that affected accounts used unsafe signing code outside its official software kits. Reused cryptographic signing nonces allowed the attacker to reconstruct API private keys from public signatures, while the protocol, contracts, infrastructure, and official kits were unaffected.
It has been revealed that the attacker created a public pool under the affected account, acquired shares through another account, then moved victim funds into the pool without creating corresponding shares. Redeeming the existing shares allowed funds to leave.
The account remarked wallet signatures were now required for public-pool transfers and that API signatures had been made non-public, with IP allowlists and configurable limits added as further protections.
Separately, Bitquery traced a 10,393,057 USDC withdrawal from Lighter's Ethereum contract on the same day. Its investigation could not inspect the internal rollup transactions or independently establish the reported second account loss.
Industry observers note that hyperliquid's API-wallet documentation notes a master account can authorize agents to sign for itself or its subaccounts. Its official agent example describes trading authority without transfer or withdrawal permissions.
The exchange endpoint allows an agent-signed asset transfer when the destination matches the source address, an expressly permitted internal movement distinct from sending assets to another owner.
According to recent reports, hyperliquid also documents a vault-transfer function, and its official Python SDK uses a different signing path for it than for external USDC transfers and bridge withdrawals. The signing code alone does not establish which agent the exchange will authorize to request the exchange.
According to recent reports, the derivatives venue's current overview describes customizable HyperEVM accounting and agent delegation, separately from legacy HyperCore vaults. Those distinctions prevent a blanket conclusion about every vault.
In a recent development, A comparable Lighter extraction would require both the relevant permissions and a matching share-accounting path. The documented internal transfers and vault functions leave no basis for a blanket assurance that every agent-authorized vault path is impossible.
A stolen agent with valid authority could open unwanted positions, change exposure, or cause adverse execution. The resulting losses could exhaust the affected account's collateral through trading and liquidation, without a withdrawal to the thief.
The Nov. 16 snapshot highlighted Hyperliquid open interest of $5.7 billion, against nearly $554 million for Lighter and a tracked total of roughly $10.5 billion. Hyperliquid was about 10.3 times Lighter's size on that measure and accounted for approximately 54.7% of the total.
Industry observers note that defiLlama's Hyperliquid and Lighter protocol pages reported the following activity and bridge snapshots on Nov. 16:.
The first layer of a hypothetical agent compromise is the authorized account. The next layer appears if that account supplies substantial market liquidity: missing quotes or forced hedging could worsen execution for others. HLP losses would require relevant strategy or liquidation exposure.
Market depth also matters more than a venue-wide balance. DefiLlama's exchange snapshot highlighted about $1.4 billion of aggregate dollar-perpetual depth within 2%, with far less available in individual markets.
Cross-venue effects could develop through shared intermediaries, hedges, and collateral needs, but those connections would have to be established for the affected participants. Lighter's incident does not demonstrate that chain of events at Hyperliquid.
In a recent development, hyperliquid holds about 55% of tracked open interest. The reach of a failure depends on the authority and exposure involved.
Illustrative: an agent with valid trading authority could cause adverse execution, unwanted positions or liquidation in its authorized account.
Illustrative: disruption to a significant provider's quotes or hedges could affect others. HLP losses require relevant exposure.
In a recent development, A separate hypothetical: settlement or withdrawal disruption could affect capital access. Cross-venue stress would require connected exposures.
Industry observers note that open interest demonstrates activity concentration. A trading-key loss, an impaired liquidity provider and a settlement failure have different paths.
The dataset is not all global crypto derivatives. Scenarios are illustrative, not observed Hyperliquid incidents. Wider effects require the stated exposure links; arrows carry no probabilities or numerical loss forecasts.
OI geometry uses the retained November 16, 2026 ranking: 5.715 / 10.456 = 54.6576%, displayed as 54.7%. The bar uses a common zero-to-total width. Scenario arrows express conditional mechanisms from documented permissions and HLP functions, with no probabilities or loss forecasts.
Sources indicate that the dataset is not all global crypto derivatives. Scenarios are illustrative, not observed Hyperliquid incidents. Wider effects require the stated exposure links; arrows carry no probabilities or numerical loss forecasts.
Hyperliquid's earlier episodes illustrate why a trader's loss, a vault drawdown and a platform interruption need separate treatment.
On March 12, 2025, approximately $4 million in HLP losses followed a large ETH liquidation, leading Hyperliquid to subsequently reduce maximum Bitcoin and Ethereum leverage to 40 times and 25 times.
It has been revealed that the March 26 JELLY episode led to a reported loss of nearly $12 million and a later operator statement putting 24-hour HLP profit at roughly 700,000 USDC. As a response, validators delisted the market after suspicious activity.
There was also a November 2025 POPCAT incident that culminated in an estimated $4.9 million HLP loss and a temporary Arbitrum-bridge pause. The fourth episode happened on April 9, with an approximately $1.5 million daily HLP drawdown amid suspected FARTCOIN manipulation.
It has been revealed that HLP's The clearest documented compensation process followed the July 29, 2025 API disruption. Hyperliquid's Aug. 4 announcement described 37 minutes of delayed orders and error responses even as transactions reached blocks, and its refund method covered specified fills and unusually high funding payments. The process excluded the smallest calculated refunds and required business or identity verification for larger recipients to receive their full balance. An Aug. 5 enhancement remarked refunds had been distributed, describing a conditional response to one operational incident. Lighter's token LIT weakened during the incident, dropping from $2.75 to $2.32 right after it, but recovering to $2.88 as of press time A historical HYPE comparison from 2025 reveals the token fell from $14.04 to $12.84 during the March 12 ETH-whale episode, before some recovery. Relevant escalation signals for Hyperliquid would include an affected account's trading authority and positions, deterioration in the markets it supports, losses reaching HLP, or disruption to settlement and withdrawals. Its $5.7 billion in open bets makes those distinctions consequential, but the dollar amount threatened by a particular compromise would still depend on the permissions, exposure, and liquidity involved. Gino Matos is a law school graduate and a seasoned journalist with six years of experience in the crypto industry. His expertise primarily focuses on the Brazilian blockchain… Hyperliquid’s $3.4 billion HYPE burn machine faces newly introduced test as USDC reserves shrink Lighter loses $15.5M in sudden drain and blames a single API leak NYSE owner takes Trade[XYZ] stake as it intends US Hyperliquid futures Hyperliquid opens native lending as HYPE hits new ATH above $90.
Recent findings suggest that the clearest documented compensation process followed the July 29, 2025 API disruption. Hyperliquid's Aug. 4 announcement described 37 minutes of delayed orders and error responses even as transactions reached blocks, and its refund method covered specified fills and unusually high funding payments.
The process excluded the smallest calculated refunds and required business or identity verification for larger recipients to receive their full balance.
An Aug. 5 enhancement remarked refunds had been distributed, describing a conditional response to one operational incident.
Lighter's token LIT weakened during the incident, dropping from $2.75 to $2.32 right after it, but recovering to $2.88 as of press time.
A historical HYPE comparison from 2025 reveals the token fell from $14.04 to $12.84 during the March 12 ETH-whale episode, before some recovery.
Relevant escalation signals for Hyperliquid would include an affected account's trading authority and positions, deterioration in the markets it supports, losses reaching HLP, or disruption to settlement and withdrawals.
Its $5.7 billion in open bets makes those distinctions consequential, but the dollar amount threatened by a particular compromise would still depend on the permissions, exposure, and liquidity involved.
Recent findings suggest that gino Matos is a law school graduate and a seasoned journalist with six years of experience in the crypto industry. His expertise primarily focuses on the Brazilian blockchain….
