Engineering

AERE goes intent-based, Across SpokePool + ERC-7683 live

By AERE Network May 31, 2026 6 min read

On May 31, 2026 AERE Network deployed two more contracts that finish the cross-chain stack: AereSpokePool (Across Protocol v3 compatible) and AereERC7683 (ERC-7683 IOriginSettler). Together they convert AERE's bridge model from "user picks chain + asset + waits 10 minutes" to "user signs one intent and a solver delivers in ~30 seconds."

What this changes: Tier 1.4 (AereMessenger) gave AERE a Hyperlane-compatible message bus. Tier 1.5 puts an intent-based UI on top of it, the same UX every modern crypto app converged on. Users no longer see "bridge." They see "deposit from any chain." Behind the scenes a competitive solver network races to fill their order.

The two contracts

Update, 2026-08-01. Both addresses in the table below are the V1 contracts, and both are now marked deprecated and flawed in our canonical registry: the settlement path could release locked principal without properly verifying the delivered output leg. Do not integrate against them. Measured on 2026-08-01, both hold a zero balance. Corrected replacements are deployed, AereSpokePoolV2Corrected 0xbEF9BF0D22dd00DBEdA262C3Cd3178f08C623d8d and AereERC7683V2Corrected 0x8eC4e7F01790C04D450b45d399E71DFd381560e3, and both are deliberately inert with no remote spoke enrolled, so no intent route is open on either version today. The mechanism described below is still an accurate description of the design; the addresses are not the ones to use.

ContractAddressWhat it does
AereSpokePool 0xCAB1DBA5f6F06198000C20a974d675f1B3181AbD Across v3-compatible SpokePool. Three core flows:
  • deposit(...), origin side. User locks input tokens, emits a V3FundsDeposited event with the intent parameters. Solvers see the event and race to fill on the destination chain.
  • fillRelay(...), destination side. Solver transfers output tokens directly to the user's recipient, then triggers a settlement message via AereMessenger.
  • handle(...), settlement. AereMessenger delivers the cross-chain receipt; the SpokePool credits the solver's pending balance, which they withdraw via claimSettlement(...).
Plus claimRefund(...) if the fill deadline passes without delivery.
AereERC7683 0x67Fb9830e3a2BC06cEb641cfF3beD87b273ccb29 ERC-7683 IOriginSettler. Implements both the on-chain (open) and gasless (openFor with EIP-712 signature) flows from the standard. Any intent aggregator that speaks ERC-7683, UniswapX, Across UI, CoW Protocol, 1inch Fusion, Anoma, can submit cross-chain orders to AERE through a uniform interface. Resolves orders into the standard ResolvedCrossChainOrder shape with maxSpent/minReceived/fillInstructions arrays.

How a consumer-app user experiences this (after solver onboarding)

Before Tier 1.5:

  1. User on Polygon wants to deposit USDC into a consumer app on AERE.
  2. The app redirects them to a bridge page.
  3. User approves USDC, locks it on the Polygon bridge contract, ~$0.30 gas.
  4. Waits 5-15 minutes for a relayer to deliver to AERE.
  5. USDC.aere arrives in their wallet, needs another step to land in their app account.

Total: 5 steps, 5-15 minutes, mental overhead at every step.

After Tier 1.5:

  1. User opens the app → "Deposit from any chain."
  2. "I have 500 USDC on Polygon." Face ID confirms (Fusaka RIP-7951 → Pectra EIP-7702 → smart account).
  3. ~30 seconds later: 499.75 USDC in their app account. The 0.25 USDC went to the solver.

Total: 1 step, 30 seconds. The bridge is invisible.

Why this is permissionless

Solvers make money on the spread between what the user pays on the source chain and what the user receives on the destination. Typical 0.05% on stablecoins, 0.1% on volatile assets. Multiplied across hundreds of millions of dollars per quarter, that's a real business, across the Across solver network, top solvers earn ~$1-5M/year per operator.

For AERE, this means: solvers compete to serve AERE. We don't recruit them, and we don't operate any solver infrastructure ourselves. Once AERE is added to the Across Protocol registry, the existing 30+ solver networks running on Ethereum/Arbitrum/Polygon/Base/Optimism/Linea/Scroll/Mantle/etc. start filling AERE intents automatically.

A consumer app also benefits: it doesn't have to build a bridge, doesn't have to manage liquidity on multiple chains, doesn't have to seed solvers. It just calls AereERC7683.open(...) with the user's signed intent.

The full Tier 1.1 → 1.5 user flow

With all five tiers shipped today, a consumer-app user's complete onboarding is:

  1. Tier 1.2 (Fusaka): User signs up with Face ID. The Apple Secure Enclave produces a P-256 signature. AERE verifies it via the RIP-7951 precompile.
  2. Tier 1.1 (Pectra): The app's smart wallet uses EIP-7702 to give the user's EOA smart-account capabilities, batched txs, social recovery, session keys.
  3. Tier 1.3 (Paymasters): The user's first 3 transactions are sponsored by AereOnboardingPaymaster. After that, they pay gas in USDC via AereTokenPaymaster, or stake AERE for a daily free-tx quota via AereStakeQuotaPaymaster.
  4. Tier 1.4 (AereMessenger): Cross-chain settlement messages travel between chains via Hyperlane-compatible ECDSA-multisig signatures.
  5. Tier 1.5 (Intents): The user signs one intent, "I want 500 USDC on AERE", and an Across solver delivers it in ~30 seconds.

Five tiers, one day, end-to-end Revolut-grade UX on a non-custodial EVM L1.

What's still pending for full operational rollout

  1. Across Protocol registry PR. Open a PR to across-protocol/contracts registering AERE chain 2800 with the SpokePool address. Typical merge + solver onboarding: 2-4 weeks.
  2. Hyperlane registry PR (from Tier 1.4), settlement messages from destination SpokePools come through AereMessenger; that needs us listed in Hyperlane's public chain registry. Same 1-3 day BD timeline.
  3. Foundation enrolls remote SpokePool addresses via enrollRemoteSpokePool(chainId, bytes32(spokePool)) once the canonical SpokePool addresses on Ethereum / Polygon / Arbitrum / BSC are determined.
  4. A consumer app's deposit UI wired against AereERC7683, the app team's task, on the app's schedule.

The on-chain layer is done. The off-chain coordination is the remaining work.

What's next

The cross-chain stack is now complete. Tier 1.6+ shifts to upgrades that strengthen what already exists, Pyth Network oracle integration (real-time prices for every dApp), drand verifiable randomness, sub-second block tuning, batch-auction settlement on AereSwap, dual burn (wiring AereFeeBurnVault into the gas flow).

The chain layer is invisible. The work continues at the application layer.

← Back to all posts