Endur
Rejected venues wait the longest for re-review; a rejection has to earn another look before the scheduled date.
Endur is a liquid staking protocol on Starknet: users stake STRK and receive xSTRK, which stays liquid while the stake earns. At the August 14, 2026 survey it held $6.0M across 6 pools, well under our $100M materiality line. A client position sized for an advised sleeve would be a large share of a venue that small, which is its own exit risk. Rejected on size: one practice advising 100 households moves $1M to $8M into a venue on the same research, and below $100M TVL that book becomes the exit crush. The file reopens if the protocol grows past the threshold and holds there.
- TVL sustained above $100M for 30 days
The research file
Applicability to the surveyed record
Endur documents xSTRK as a Starknet liquid-staking receipt issued by an ERC-4626 vault. Deposits first meet withdrawal demand, excess STRK is distributed through delegator contracts to validators, and compounded staking rewards increase xSTRK’s exchange value, establishing the current liquid-staking mechanism.
Current observation and perimeter
The DefiLlama protocol API read on 2026-08-15 classified Endur as Liquid Staking, reported only Starknet, and showed approximately $5.56M TVL. Endur’s current docs also list BTC-denominated LST products, but this surveyed record and its description remain xSTRK-focused; aggregate scale is far below the shared v1 $100M threshold.
Control and exit applicability
A validator registry whitelists validators, assets, and delegators, while an automated relayer manages allocations and withdrawal processing under contract constraints. xSTRK redemption issues a queue NFT and is funded by new deposits or validator unstaking; normal completion ranges from hours to eight days and can reach fourteen in extreme demand, while any instant DEX exit depends on secondary liquidity and price.
Why the class rule decides
The shared v1 below-materiality dossier controls because Endur remains under $100M despite a live, documented xSTRK system. Reopen after DefiLlama TVL remains above $100M for 30 consecutive days, then review registry and relayer authority, validator selection and concentration, contract upgrade control, Starknet staking dependencies, audits and incidents, xSTRK exchange-rate accounting, withdrawal-queue history, executable DEX liquidity, stressed exit, and named liquid-staking alternatives.
Sources
The claims above trace to these. Where a number could not be independently verified, the thesis says so.
- Endur — liquid-staking architecture · primary · accessed 2026-08-15
Supports: ERC-4626 LST, withdraw queue, delegator contracts, validator registry, deposit and withdrawal flow - Endur — xSTRK mechanism · primary · accessed 2026-08-15
Supports: xSTRK receipt, exchange rate, staking rewards, auto-compounding - Endur — withdrawal log · primary · accessed 2026-08-15
Supports: NFT queue, funding sources, normal redemption, 14-day stress case, permissionless claim - Endur — mainnet contracts · primary · accessed 2026-08-15
Supports: Starknet deployment, xSTRK contract, withdraw queue contract, validator registry contract - DefiLlama — Endur survey record · secondary · accessed 2026-08-15
Supports: current TVL, Starknet perimeter, Liquid Staking category, survey observation
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 |
|---|---|---|---|
| Starknet | Approved · limits | hybrid | validity proofs and a regular exit window constrain control, but permissioned proposers and an instant emergency Security Council remain live dependencies. |