Skip to content
PACE
Stewarded by

The Portable Account Core for Ethereum

The Hegotá upgrade adds a new account type to Ethereum, and every wallet has to rebuild for it. PACE is an industry effort to agree on the core of that account: one spec every wallet implements and a few contracts they all share. Your account then opens in any wallet, and any hardware wallet can read what you sign.

Why this needs to exist

Each time Ethereum got a new kind of account, every wallet built its own version.

  1. 2023 ERC-4337

    A separate account in every wallet

    Accounts rarely opened in another wallet.

  2. 2025 EIP-7702

    New code for old addresses

    Every wallet wrote its own again, and the account’s original key still controls everything.

  3. Hegotá EIP-8141This time

    A new account in the protocol

    Hegotá is expected in 2027, and every wallet rebuilds for it at the same moment. After it, account features ship as contracts with no fork, so nothing makes wallets rebuild together again.

If every wallet builds its own

  • Wallet A
  • Wallet B
  • Wallet C
  • Account A
  • Account B
  • Account C
  • Apps
  • Hardware wallets
  • Explorers

Apps, hardware wallets and explorers support each account separately.

If every wallet ships the same core

  • Wallet A
  • Wallet B
  • Wallet C

PACE

  • Apps
  • Hardware wallets
  • Explorers

Apps, hardware wallets and explorers support the core once.

  • Users

    Switching wallets means starting over.

    No other wallet can read your keys, so you move every asset by hand.

  • Explorers

    Explorers can’t show what an account did.

    Without a decoder for each wallet, a key change shows up as an unlabelled call.

  • Hardware wallets

    Hardware wallets show raw bytes.

    For an account the device doesn’t know, you approve hex you can’t read.

  • Security

    Transaction checks work in only some wallets.

    Assertions, the checks the chain runs on a transaction’s result, need one encoding. Otherwise hardware wallets can’t show them and apps can’t rely on them.

What we need to do

PACE is one spec plus a few shared contracts, starting with EIP-8130’s Keystore, the contract that holds every account’s keys. We build the contracts and the parts apps and hardware wallets read, and other teams ship the rest.

The work, by where it lands

Apps and tools

What apps, devices and explorers read

  1. Give apps one API for PACE

    Us, with the ERC-5792 authors

  2. Write the clear signing metadata

    for the clearsigning.org registry

    Us, with the ERC-7730 group

  3. Get explorers and security tools to read PACE

    Explorers, simulators, security tools

Contracts

Deployed once, shared by every wallet

  1. Deploy the Keystore on Ethereum

    Us · no fork

  2. Build the Ethereum account that reads the Keystore

    Us · no fork

  3. Build the threshold authenticator for multisigs and co-signers

    Us · no fork

  4. Pay gas in tokens like USDC

    least settled item

    A working group · new ERC

Network rules

Node policy, no fork

  1. Allow Keystore reads in the public mempool

    proposed in EIPs#12340

    L1 client teams

  2. Ship parallel nonces

    EIP-8250 is expected in Hegotá

    EIP-8250 authors, L1 client teams

Protocol

Ships with a fork or an L2 upgrade

  1. Get every L2 to support assertions

    EIP-7906, expected on L1 in Hegotá

    Each L2, the EIP-8130 authors

  2. Adjust EIP-8141 for app permissions

    The EIP-8141 authors

What every user gets

These features come with PACE, so they work in every wallet. The screens are from the interactive mockup.

Move to post-quantum keys

Quantum computers could break ECDSA, the signature behind every Ethereum key today. In EIP-8130 a contract verifies each key’s signature, so a post-quantum key is just another key, and you keep your address and funds.

Post-quantum keys

Add one next to your current keys. Every wallet that opens the account sees it.

Retire the old keys

Remove the ECDSA keys and keep the same address.

New schemes without a fork

Signature schemes ship as contracts, so a better one needs no hard fork.

Every chain at once

One key change applies on every chain the account is on.

Approve the result you expect

Getting drained is the biggest risk of holding your own keys. A co-signer can refuse a bad transaction before it counts, and the chain reverts any result you didn’t approve.

Transaction assertions

Your wallet simulates the transaction and turns the result into assertions. If anything else changes, the chain reverts it.

Security co-signer

A shared threshold authenticator lets your key and a security service sign as one. The service screens each transaction and can refuse.

App permissions with limits

Give an app a cap and an end date, like 50 USDC a day for a week.

Remove access from any wallet

Any wallet that opens the account can see and revoke what an app may do.

Read every transaction on your hardware wallet

A hacked web page can show one thing and ask you to sign another, which is how Bybit lost about $1.5 billion in 2025. PACE includes clear signing metadata for every operation, so the hardware wallet decodes the transaction itself and a hacked screen can’t fool you.

Every step in words

The device walks through the batch one call at a time.

Assertions on the device

Before you approve, it lists the checks, like at most 1,000 USDC out.

Same words in every wallet

Each kind of call has one shared description, so every wallet and device uses the same words.

Words for every kind of action

Adding a device or giving an app access shows in plain words, never as code.

Use Ethereum like a banking app

Passkeys, batches and sponsored gas come from EIP-8130, so they work in every wallet.

Passkey signing

Sign with Face ID. The key never leaves the phone.

One signature for the whole batch

Approve, swap and deposit in one transaction.

Someone else pays the gas

An app or a sponsor pays the fee, so a new account needs no ETH.

A stuck payment holds nothing up

Payments don’t wait in one line, so a stuck one never blocks the next.

What app builders get

Today an app builds a separate transaction path for each kind of account its users bring, and falls back whenever a wallet can’t do something. With PACE every wallet answers the same way, so the app reasons about one path.

Today: a path for each kind of account

  • Plain EOA

    Approve, then act, with one signature per call and gas paid in ETH.

  • EOA upgraded with EIP-7702

    A batch if the wallet supports it, through code each wallet chose.

  • ERC-4337 account

    A user operation through a bundler, with each wallet’s own modules.

  • Every wallet in between

    Its own answer on sponsored gas, app permissions and receipts.

Each path needs its own code, its own tests and a fallback for what a wallet can’t do.

With PACE: one path

  • Send

    One wallet_sendCalls, with phases, a gas payer and assertions.

  • Ask for access

    One ERC-7715 request, used through ERC-7710.

  • Read the result

    The same receipt on every chain, including a failed assertion.

  • Check the chain

    One wallet_getCapabilities call says what holds there.

Every wallet that ships PACE runs it the same way, including for EOAs that upgrade in place.

What a wallet can add

These aren’t part of PACE. A wallet adds them on top, and other wallets can still see them even when they can’t use them.

Spending limits

Cap what a key spends per day, week or month.

Cross-chain in one approval

Move USDC from Base to Arbitrum with one signature.

Private balance

Hold ETH in a shielded pool and pay without linking back to your account.

Keys from a recovery phrase

Rebuild your keys on a new device from a 12-word phrase.

The road to Hegotá

The goal is every wallet shipping PACE by default when Hegotá activates.

From draft to mainnet
  1. Out now

    Public draft

    The ERC and the interactive mockup.

  2. In progress

    Working group

    Wallet, client, hardware, security, explorer and EF teams.

Ethlabs builds the contracts through v1 and organizes the working group.

Help decide what goes into PACE before Hegotá.

If you build wallets, apps, hardware wallets or tools, tell us what PACE is missing or what should be an extension. Joining the working group is how you get a say.

Project
PACE: Portable Account Core for Ethereum
Number
ERC-XXXX
Status
Draft, open for comment
First draft
21 Sep 2026
Target
Hegotá
Drawn by
Ethlabs
Reviewed by
your team?

Acknowledgements

  • Keystore AccountsEIP-8130

    C. Hunter

  • Execution permissionsERC-7715

    L. Isailovic, D. Rein, D. Finlay, D. Chiang, F. Makarov,P. Gomes, C. Swenberg, L. Rosario, I. Bowman, J. Smale

  • Smart contract delegationERC-7710

    R. McPeck, D. Finlay, R. Dawson, D. Chiang

  • Transaction AssertionsEIP-7906

    A. Forshtat, S. Nacson, D. Tirosh, Y. Weiss, F. Svantes

  • Frame TransactionEIP-8141

    V. Buterin, lightclient, F. Lange, Y. Weiss, A. Forshtat,D. Tirosh, S. Nacson, D. Chiang, T. Wahrstätter, S. Vlachakis

  • Clear SigningERC-7730

    L. Castillo, D. Rein, P. Aoun, A. Galansky,B. Rozwarski, K. Uzdogan, Fredrik, A. Forshtat

  • Wallet Call APIERC-5792

    M. Salem, L. Rosario, W. Cusack, D. Tirosh, J. Moxey, D. Rein, A. Forshtat,S. Wilson, B. Itskov, J. Tavares, A. Fuller, P. Liao, bumblefudge

  • Recent RootsEIP-8272

    T. Thiery, V. Buterin, T. Wahrstätter, lightclient

  • Smart BatchingERC-8211

    M. Javor, F. Dujmusic, F. Makarov, V. Rajendran

  • Crosschain EIP-712ERC-7964

    E. Garcia

  • User operationsERC-4337

    V. Buterin, Y. Weiss, D. Tirosh,S. Nacson, A. Forshtat, K. Gazso, T. Hess