StarkGate (Starknet Bridge)

The research assessment is adverse because of a recent reliability gap that has not yet been resolved, though this is the narrowest rejection in this batch. L2Beat confirms that Starknet is a Stage 1 ZK rollup, the same rating as Base. Its governance is comparably split between a 2-of-6 StarkWare multisig that sends normal actions through an eight-day on-chain timelock and a 9-of-12 Security Council that can upgrade instantly only in emergencies. Base, Arbitrum, and Optimism have the same zero-delay-once-invoked structure reflected in their registry caps. The incident record keeps Starknet at zero: mainnet halted twice in four months, in September 2025 and on 2026-01-05. Neither halt lost user funds, but both stopped block production. At this review, roughly seven clean months have passed since the later halt, short of the twelve-month bar this registry applies elsewhere. This is the bridges batch’s closest case for a near-term reopen, not a structural disqualification.

Review triggers: events that reopen the verdict
• Twelve consecutive months with no mainnet halt or block-production outage, counted from the 2026-01-05 incident
• Any future major protocol upgrade is accompanied by a public pre-mortem or rollback plan, reducing recurrence risk
• StarkWare Multisig or Security Council composition or threshold changes are publicly disclosed within 48 hours
• Any confirmed exploit of the StarkGate bridge contracts themselves

Watched nightly: a warning on its venues or files, or a cited document that changes, reopens the memo. The first confirmation is due 2026-11-17.

The research file

Mechanism

Starknet is a Cairo-based ZK validity rollup. STARK proofs prove state transitions instead of leaving them assumed unless someone disputes them. The L1 bridge contract unlocks withdrawn funds only after it verifies the STARK proof that covers the corresponding L2 burn. This removes the fraud-proof challenge period. Once L1 proves and accepts the relevant block, withdrawals typically clear in hours instead of the seven days required by OP Stack chains.

Control and governance

L2Beat directly confirms that Starknet is rated Stage 1. Control of the main rollup contract and its ETHBridge and STRKBridge escrows is split between a 2-of-6 StarkWare Multisig and a 9-of-12 Security Council. The multisig sends all normal actions through a DelayedExecutor timelock with an eight-day on-chain delay. The Security Council can upgrade instantly only in emergencies. A separate 2-of-4 StarkWare Multisig 2 controls most other bridge escrows and can upgrade them instantly. The permissioned Operator role supplies state-transition proofs. To resist censorship, Starknet uses a StarknetSCMinorityMultisig that lets a 3-of-12 Security Council minority include transactions that regular Operators have not included. All bridge escrows can turn on a withdrawal throttle that caps flows at 5% of locked funds per 24-hour period. A multisig has permission to turn it on. In the core bridges, only a 3-of-12 Security Council minority is needed to turn it off. L2Beat says Starknet does not pass its ”walkaway test”: users cannot exit if the operator is malicious and the Security Council itself disappears. Base does not have this gap, though it is narrower than the CRITICAL, no-delay upgrade paths that caused several other entries in this batch to fail.

Incident record

Starknet had two confirmed mainnet halts within roughly four months. A nine-hour block-production halt in September 2025 followed the Grinta (v0.14.0) upgrade combined with Ethereum RPC provider failures and sequencer bugs. A four-hour halt on 2026-01-05 came from a proving error on a pending transaction, after which engineers paused sequencers as a precaution. Neither event lost user funds, and we found no exploit of the StarkGate bridge contracts themselves. Still, two significant halts in four months show a reliability pattern absent from the records of Base, Arbitrum, and Optimism’s canonical bridges. As of this review, only about seven clean months have passed since the later incident.

Exit under stress

Normal withdrawals are fast. That is a real structural advantage of the validity-proof design over every optimistic bridge in this registry. But a sequencer halt, which happened twice in the history reviewed, stops the production of all new blocks and proofs. Withdrawal finality therefore depends on sequencer uptime, which has a recent, documented gap.

Comparison

Starknet’s split multisig and Security Council structure, with an eight-day delay for non-emergency actions, is comparable to Arbitrum’s approved structure and arguably somewhat more conservative. Starknet also discloses its controls far better than World Chain or Robinhood Chain, which have verified weaknesses elsewhere in this batch. This is not a Linea- or Mantle-style ”None” exit-window case. The only material difference from the three approved peers is the incident record: Starknet has a real, recent reliability gap that Base, Arbitrum, and Optimism have not shown. Its governance structure is not the reason for rejection.

Sources

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

• L2Beat — Starknet · primary · accessed 2026-08-17
  Supports: Stage 1 rating, 2-of-6 multisig plus 8-day timelock, 9-of-12 Security Council emergency path, fails walkaway test, withdrawal throttle mechanics
• Starknet — official incident report, January 5, 2026 · primary · accessed 2026-08-17
  Supports: January 2026 halt, proving error root cause
• The Block — Starknet outage marks second major disruption since Grinta rollout · secondary · accessed 2026-08-17
  Supports: links January halt to September 2025 Grinta incident
• Starknet Documentation — StarkGate · primary · accessed 2026-08-17
  Supports: bridge mechanism, validity-proof-gated withdrawal
• CoinDesk — Starknet goes down for over 4 hours · secondary · accessed 2026-08-17
  Supports: January 2026 outage timeline

Inherited controls

The research above describes the protocol layer. Every position also inherits the asset it holds and the chain it settles on. The layer with the most administrative power sets the position’s effective control; that describes control, not quality or suitability.