GILDer

Rejected venues wait the longest for re-review; a rejection has to earn another look before the scheduled date.

GILDer is rejected under the version-1 below-materiality dossier. The new feed entry held about $16,800 and advertised a 20% base APY, more than 5,900 times below the $100M protocol-capacity screen. The upstream record describes a three-year USDC term deposit represented by an NFT, with 80% held in a per-user Safe and the remainder funding an unspecified yield engine. This is a protocol-specific application of a published class rule, not an unsupported claim that every contract or operator behind GILDer is defective.

Kill criteria: the events that revoke this verdict
• GILDer sustains at least $100 million of reconciled protocol TVL for 30 consecutive days
• GILDer publishes its operator, legal terms, audits, yield-engine assets and counterparties, and an executable early-exit process

The research file

Mechanism and applicability

The upstream record describes a three-year USDC term deposit represented by an NFT, with 80% held in a per-user Safe and the remainder funding an unspecified yield engine. The reviewed deployment therefore satisfies the dossier’s mechanism requirements on its own facts. The new feed entry held about $16,800 and advertised a 20% base APY, more than 5,900 times below the $100M protocol-capacity screen. A class application records enough protocol evidence to prove applicability while leaving the shared economic argument in the pinned dossier rather than pretending this is a separate flagship review.

Control and incident boundary

DefiLlama supplied no protocol website, named operator, audit, legal terms, or disclosed yield-engine counterparties; the class rule stops on capacity before those gaps could be underwritten. Those controls and the available incident record may change operational risk, but they do not remove the property that triggers this disposition. No clean-record claim is used as proof of safety: a young deployment can have little adversarial history, and an established deployment can execute its intended economics without an exploit while still remaining unsuitable for the advised sleeve.

Exit and current measurement

A three-year term and an NFT claim are not equivalent to executable USDC liquidity, and the retained 20% yield-engine sleeve adds a separate recovery dependency. Aggregate TVL is an accounting measure rather than a promise that the exact client position can be unwound at the displayed value. The disposition remains a zero allocation until a stated reopen condition is observed and a new review measures the proposed-size exit instead of inferring it from a dashboard total.

Comparison and decision

Material reviewed stablecoin venues provide deeper exits, shorter review cycles, and disclosed mechanisms without asking one practice to become dominant capital. The comparison is made at the exposure level, not by brand or headline rate. The published dossier is preferable to repeating the same class judgment with slightly different wording for every venue; the protocol-specific sources retained here make the classification reproducible and the reopen criteria observable.

Sources

The claims above trace to these. Where a number could not be independently verified, the thesis says so.

• Safe Documentation — smart accounts · primary · accessed 2026-08-19
  Supports: Safe account control boundary, signer model
• DefiLlama — GILDer protocol data · secondary · accessed 2026-08-19
  Supports: protocol category, chain perimeter, current TVL

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.