Cross-chain DeFi exists because useful assets, pools, and strategies do not all live on one chain.
When users move capital across chains through bridge architecture, they are not just adding one more transaction. They are adding another contract layer, another failure surface, and another place where things can go wrong before the strategy even begins.
This article looks at where those risks actually sit, what they have cost users in practice, and why resolver-based HTLC architecture changes the picture. Omniston, STON.fi’s cross-chain execution layer, gives users a different way to reach supported EVM chains without carrying the usual bridge-architecture stack into the route.
Quick highlights
- Bridge contracts hold concentrated locked assets, which makes them one of the biggest exploit targets in DeFi.
- A bridge move usually adds four cost layers: origin gas, bridge fee, destination gas, and swap slippage after arrival.
- Wrapped tokens are only as good as the bridge contract backing them.
- Validator or signer compromise has caused some of the biggest bridge losses in crypto.
- Omniston uses resolver-based HTLC settlement instead of a shared bridge contract, with three valid outcomes and no path in which both parties lose funds.
- Omniston is stablecoin-first by design, which fits the part of cross-chain activity where users and apps care most about predictability: deposits, payments, treasury movement, exchange flows, app access, and automated execution.
- It also changes the user experience in two practical ways: the cross-chain leg settles faster than a typical bridge route, and the quote shown at confirmation is the price that executes.
What “cross-chain” means for assets and liquidity in crypto
Cross-chain DeFi means moving tokens from one blockchain to another to reach assets, pools, or protocols that do not exist on the origin chain.
There are two broad ways to do that.
- The traditional route is a bridge. Assets are locked on Chain A, and a wrapped representation appears on Chain B. The original asset does not move anywhere. It sits in a contract while the destination chain receives what is effectively a claim on that locked balance.
- The other route is a resolver-based HTLC system. Instead of minting a wrapped token, it uses paired Hashed Timelock Contracts on both chains and a competing resolver market to swap the source asset into the native destination asset directly.
Liquidity in crypto fragments naturally across blockchains because each ecosystem runs its own pools and incentive programs. A pool deep enough to handle a $500,000 swap on Ethereum may have no equivalent on a newer chain — and vice versa. Cross-chain strategies look attractive precisely because of that gap, which is also what makes them expensive to execute without accumulating meaningful risk along the way, in either architecture.
How cross-chain bridge security risk accumulates
Bridge risk isn’t a single event — it’s a stack of vulnerabilities that compound with every additional protocol interaction.
Smart contract vulnerability: why bridge contracts attract the biggest attacks
A bridge contract does something unusual in DeFi: it holds a large amount of value at rest.
That makes it an attractive target. In 2022 alone, roughly $2 billion was stolen across 13 cross-chain bridge hacks, representing about 69% of all crypto funds stolen that year. The pattern is simple enough: when a system concentrates a large amount of idle collateral behind one contract layer, attackers notice.
Custodial and validator risk between chains
Many bridges rely on a validator set or multisig committee to approve actions between chains.
That creates a point of concentration even when the system is presented as decentralized. Some of the largest bridge losses did not come from subtle code bugs alone. They came from compromise at the signing layer. Ronin and Orbit Chain are the clearest examples. When the keys controlling the route are compromised, the assets locked on the origin side are what end up exposed.
The Ronin Bridge was drained of approximately $625 million in March 2022 after attackers compromised five of nine validator keys.
The Orbit Chain bridge lost $81 million in January 2024 through a parallel mechanism: compromise of a 7-of-10 signing committee.
Fee and timing risk: what a cross-chain bridge move actually costs
Even when a bridge works exactly as intended, it still changes the economics of the move.
A cross-chain bridge route usually adds origin-chain gas, bridge fee, destination-chain gas, and swap slippage after arrival. Each one can look manageable in isolation. Together, they can erase much of the yield advantage that made the move look attractive in the first place.
Time adds another layer. Bridge routes often take several minutes, and sometimes far longer, for destination assets to arrive. That delay is structural, not unusual. By the time the transfer clears, the pool, price, or yield opportunity may already have changed.
When cross-chain economics fail: scenarios that erode the advantage
- Scenario 1 — the yield opportunity that disappears in transit. A user sees a stronger yield on another chain, starts the bridge, and waits. The route takes longer than expected. By the time the assets arrive, more capital has already moved into the same opportunity and the APY has compressed. After fees and slippage, the move no longer looks smart.
- Scenario 2 — the wrapped token that loses its peg. A user lands on the destination chain holding a wrapped representation that depends on the bridge contract behind it. If that bridge later suffers an exploit or message-validation failure, the wrapped asset can lose its practical reliability even if the user never interacted with the exploit directly.
The Kelp DAO exploit in April 2026 drained approximately $292 million in rsETH via a forged cross-chain message, compromising backing across more than twenty chains simultaneously. - Scenario 3 — liquidity fragmentation at scale. A user attempts to swap a $50,000 position, and then discovers that the destination pool cannot absorb the size without real price impact. The bridge cost is already paid by the time this becomes obvious.
Capital in a multi-chain crypto strategy is dispersed across thousands of isolated pools, so a single DEX on a smaller chain often can’t absorb large swaps without significant price impact.
A newer architectural option: resolver-based HTLC via Omniston
The usual framing is: either bridge across chains and accept the bridge risk stack, or stay native and accept the limits of a single ecosystem. But there is a third architecture.
Omniston, STON.fi’s cross-chain execution layer, uses paired HTLCs and a competing RFQ market of professional resolvers to deliver native destination assets directly.
The flow is straightforward. The user signs a quote request. Resolvers compete to fill it. The winning resolver locks destination-side assets in an HTLC on the destination chain while the user’s funds lock in a paired HTLC on the source chain. When the secret is revealed, both legs settle atomically.
That changes several things at once.
- First, there is no shared bridge contract holding pooled collateral in the middle of the route.
- Second, there is no wrapped representation created on arrival. The user receives the native destination asset.
- Third, the failure logic is narrower and cleaner. Omniston is built around three valid outcomes: both parties receive what they were quoted, the user is refunded by timelock if the resolver fails to respond, or the resolver is refunded if the secret is never disclosed. There is no valid outcome in which both parties lose funds. That is what “all-or-nothing” means in practice.
Why Omniston adds real credibility
- Omniston is already a battle-tested liquidity aggregation protocol and RFQ layer with real adoption, real user activity, and live use across the TON ecosystem. The execution layer is already trusted in real conditions.
- Omniston is stablecoin-first by design. That is exactly where many serious cross-chain use cases live. Users and apps care most about predictability when the goal is clean value movement rather than token speculation. That is why stablecoin routing is such an important proving ground. It is where users notice inconsistency fastest, and where clean execution matters most.
- Omniston offers faster settlement. Bridge routes often take several minutes, and sometimes much longer, for destination assets to arrive. Omniston does not insert that extra bridge-confirmation phase in the middle. The cross-chain leg settles through paired HTLCs as soon as the resolver fills the quote, so the destination side completes in a single execution window rather than after a long bridge delay.
- Omniston guarantees price certainty at confirmation. The resolver-signed quote is the price that executes. There is no gap between what the user approves and what later appears on-chain. What you see is what you get. For cross-chain users, that is a bigger improvement than it sounds. A lot of frustration comes from signing one number and receiving another.
Cross-chain via bridge, native single-chain, and resolver-based HTLC: a side-by-side view
| Dimension | Cross-chain via bridge | Native single-chain DeFi | Resolver-based HTLC via Omniston (STON.fi’s cross-chain execution layer) |
| Asset custody | Assets locked in a bridge contract during transit | Assets remain in user’s non-custodial wallet throughout | Source-side and destination-side HTLCs hold both legs in cryptographic escrow; refundable by timelock |
| Smart-contract exposure | Origin bridge + destination bridge + destination DEX | Single-chain contracts only; no bridge interaction | Paired HTLCs per swap; no shared bridge contract; no validator multi-sig holding keys |
| Fee structure | Origin gas + bridge fee + destination gas + swap slippage | One gas fee + DEX swap fee | Source-chain gas + RFQ-competitive resolver margin + destination-chain gas |
| Failure recovery | Manual claim, sometimes a support ticket; outcomes vary | Single-chain transaction reverts instantly if it fails | Three outcomes guaranteed: both paid, user refunded by timelock, or resolver refunded by timelock — never both parties losing funds |
| Liquidity depth | Fragmented across chains; thinner pools on newer networks | Concentrated in native pools; deeper for ecosystem tokens | Native depth on each chain; resolver network sources liquidity competitively, not from parallel deployments |
| Token access at destination | Wrapped IOU dependent on origin-chain contract solvency | Limited to tokens native to the home chain only | Native destination asset, no wrapped representation anywhere in the route |
| Best for | Routes where no resolver-based alternative exists | Maximizing security and fee efficiency within one ecosystem | Cross-chain access into Phase 1 EVM destinations (Ethereum, BNB Chain, Base, Polygon) without bridge-architecture risk |
Five checks before any cross-chain move
Before sending funds across chains, ask a few simple questions.
- What is the full end-to-end cost of the move?
- How deep is the destination pool at the size I actually plan to use?
- Do I need a native destination asset, or am I accepting wrapped-token dependency?
- If I am using a bridge, who controls the signing layer and how recent is the audit?
- Is the opportunity still worth it after route cost, timing risk, and possible slippage are included?
Wrapping up
Cross-chain DeFi does offer real opportunities. The problem is that many users still reach them through an architecture that adds risk, delay, and extra cost before the strategy even starts.
That is the whole stack of issues with bridge-based movement: concentrated locked value, signer-layer risk, wrapped-token dependency, timing drag, and fee drag.
For users working inside TON, STON.fi already handles the native intrachain side directly. For users moving into supported EVM destinations, Omniston provides a different cross-chain path: native destination assets, HTLC-based settlement, faster completion, and quote certainty at the confirm screen, without relying on shared bridge architecture.
