Spark Liquidity Layer
Rejected venues wait the longest for re-review; a rejection has to earn another look before the scheduled date.
REJECTED ON UNDISCLOSED LEGAL AND CROSS-CHAIN ADMIN STRUCTURE. Spark Liquidity Layer is the capital-allocation plumbing beneath Spark Savings’ sUSDS product, routing USDS, sUSDS, and USDC across lending markets, ERC-4626 vaults, DEXs, and staking venues on eight chains under Sky ecosystem governance. Despite the eight-chain footprint, roughly 96% of tracked capital sits on Ethereum mainnet today, meaning the cross-chain admin surface is currently more a governance exposure than a large realized capital one — a genuine, favorable finding this review credits. But this review could not confirm the operating legal entity’s name or jurisdiction, any KYC or geographic eligibility policy, or how the protocol’s admin, relayer, and freezer roles are actually structured chain by chain — whether each of the eight deployments has its own local pause authority or all control flows from Ethereum mainnet governance via cross-chain messaging is undisclosed in any source this review could access.
- The operating legal entity’s name and incorporation jurisdiction are publicly disclosed
- Per-chain admin, relayer, and freezer authority structure is disclosed, including whether each deployment has independent local control
- A specific KYC and geographic eligibility policy is published
- Non-Ethereum chain TVL concentration is monitored, with this entry reopened for reassessment if a non-Ethereum deployment grows to material size before the admin-structure gap is closed
The research file
Mechanism
Spark Liquidity Layer is described in Spark’s own documentation as a non-custodial capital allocator operating through governance-approved venues with rate limits meant to keep capital movement ”constrained, predictable, and bounded.” Allocation venues include SparkLend and other Aave-compatible markets, Morpho vaults curated by Spark, Curve and Uniswap v4 stableswap pools, and staking positions in Ethena USDe, Lido wstETH, and EtherFi weETH. A depositor into Spark Savings receives sUSDS, an ERC-4626 share accruing the Sky Savings Rate — a rate Spark cannot set unilaterally, since it is funded by Sky protocol borrowing fees and set by Sky governance.
Concentration reduces current cross-chain exposure
Current chain-by-chain TVL: Ethereum roughly $1.69B (about 96% of the total), Base roughly $64M, Arbitrum roughly $10M, and the remaining five chains (Optimism, Unichain, Robinhood Chain, Avalanche, X Layer) combined under $2M, several at or near zero. Despite governance and code supporting an eight-chain deployment, actual capital exposure to non-Ethereum bridge and admin-key risk is currently small in dollar terms — a real mitigating fact, though not a resolution of the underlying disclosure gap below, since concentration can shift as chains scale.
Undisclosed legal entity and eligibility
This review could not confirm Spark’s operating legal entity name or incorporation jurisdiction from any fetchable source — Spark’s legal and terms-of-service pages returned not-found errors, and the app’s footer could not be independently retrieved. No KYC policy or geographic access restriction was found in Spark’s FAQ or product documentation; deposits appear to occur through the standard app interface without documented gating, but this is an inference from absence of a stated policy, not a confirmed fact.
Admin structure across chains
The Spark ALM Controller’s access-control model documents a `DEFAULT_ADMIN_ROLE` run by governance, a `RELAYER` role assumed to be potentially compromisable and constrained by rate limits, and a `FREEZER` role that can halt operations by removing a compromised relayer — a reasonable design on paper. But whether this structure is independently deployed and controlled per chain, or centrally directed from Ethereum mainnet governance via cross-chain messaging, could not be confirmed; direct on-chain contract-owner verification via block explorers was blocked in this research pass. Nor could the specific bridge mechanism used for cross-chain transfers on each deployment be confirmed.
Redemption, audits, and comparison
On Ethereum mainnet, the sUSDS vault holds a liquidity buffer of up to $10M for atomic redemptions, with larger withdrawals settled through an asynchronous intent mechanism typically fulfilled within minutes; Spark’s own documentation confirms that on other chains, deposits and withdrawals depend on the Liquidity Layer routing capital across networks, meaning local liquidity is not always pre-funded and depends on rebalancing working correctly. Multiple audit rounds from Cantina, ChainSecurity, and Certora cover the controller contracts across several versions — a genuine security investment. Against Grove, another Sky-ecosystem allocator researched alongside this entry and also rejected, Spark Liquidity Layer is the narrower plumbing layer underneath an existing savings product rather than a distinct credit-strategy allocator, but carries the same undisclosed-entity gap.
Sources
The claims above trace to these. Where a number could not be independently verified, the thesis says so.
- Spark documentation — homepage and Liquidity Layer overview · primary · accessed 2026-08-19
Supports: allocation venues, non-custodial capital allocator framing - Spark documentation — Spark Savings product page · primary · accessed 2026-08-19
Supports: sUSDS mechanics, Ethereum liquidity buffer size, Sky Savings Rate governance - Spark documentation — governance overview · primary · accessed 2026-08-19
Supports: Spark Risk Council, Operational Facilitator, Sky Atlas spell process - Spark ALM Controller — GitHub repository and audits · primary · accessed 2026-08-19
Supports: admin, relayer, and freezer role design, Cantina, ChainSecurity, and Certora audit history - DefiLlama — Spark Liquidity Layer protocol data · secondary · accessed 2026-08-19
Supports: current chain-by-chain TVL breakdown
Inherited controls
The verdict above grades the protocol layer. Every position also inherits the asset it holds and the chain it settles on. The least safe layer sets the position’s grade, and the position table names which one that is.
| Chain | Verdict | Grade | Control constraint |
|---|---|---|---|
| Ethereum | Approved | sovereign | No sequencer, no upgrade key, no operator who can be compelled — rule changes require social consensus. |
| Base | Approved · limits | hybrid | Coinbase — one regulated US company — operates the only sequencer, and admin keys can upgrade bridge contracts within ~7 days. |
| Arbitrum One | Approved · limits | hybrid | a single sequencer orders >99% of transactions and admin keys can upgrade bridge contracts on a ~7-day timelock. |
| OP Mainnet | Rejected | hybrid | Ethereum forced inclusion limits sequencer censorship, but the Foundation and Security Council can co-sign an immediate upgrade before a client can exit. |
| Avalanche | Approved · limits | crypto-backed | no party can freeze or seize C-Chain funds, but one vendor writes the only production client and Messari measured over a third of stake hosted on AWS. |
| Robinhood Chain | Rejected | hybrid | one sequencer and two permissioned validators sit beneath an emergency council and transaction filter that can defeat the normal force-inclusion backstop. |
| X Layer | Rejected | freezable | OKX operates the ordering path, proof roles are permissioned, and an X Layer multisig can upgrade immediately; the operator has also suspended block production for an upgrade. |