AA2026-09-08_004

Native Account Abstraction #004

2026-09-08 3 decisions 387 transcript lines

Transcript

Follow along

marc | wolovim
2 min to start
marc | wolovim
agenda: https://github.com/ethereum/pm/issues/2212
marc | wolovim
PRs against 8141: https://forkcast.org/eips/8141/?tab=history
Guru
https://github.com/ethereum/execution-specs/releases#release-tests-frames-devnet@v0.3.0
lightclient
i dont
Marius van der Wijden
BB but he's on vaca I think
lightclient
Reacted to "BB but he's on vaca I think" with 🔥
lightclient
geth branch is also frames-devnet-0
Guru
Reacted to "geth branch is also ..." with 👍
Nico
Gm AA frens
lightclient
that would be cool!
Alex Forshtat
I would like to ask for some feedback on EIP-7906 and the general direction of transaction assertions, post-transaction frame mode and journal/diff access opcodes. This was posted as the protocol review of the proposal: "Proposed narrowing pending further research: arbitrary storage reads removed, assertions over emitted events and BAL-touched slots deliver the value." It would be great to hear what the BAL inspired opcode format would have to look like in your opinion.
lightclient
this complicates the core protocol though 🙂
lightclient
the VERIFY calldata is signed though
soispoke
we went through the same discusions for recent roots btw (8272)
soispoke
Répondre à "we went through th..."
txn enveloppe field
//github.com/ethereum/EIPs/pull/12281
Daniil Ankushin | Nethermind
https://github.com/ethereum/ERCs/pull/1979 https://excalidraw.com/#json=2pWxeBerW4Fq8GW-LRku0,FfVXlml_9qxztvasPFKxLQ
Daniil Ankushin | Nethermind
https://github.com/ethereum/EIPs/pull/12301
Iván | ethrex
I’ll take a look at it!
raxhvl
Replying to "this complicates t..." curious why though? doesnt expiry, like signature, is a validation the apply to the whole transaction and fits well at the top level? this is a pure comparison - adding a state when dont really need it is an overkill no?
Daniil Ankushin | Nethermind
Reacted to "I’ll take a look at it!" with ❤️
Iván | ethrex
https://faucet.frames.ethrex.xyz
marc | wolovim
7906: https://eips.ethereum.org/EIPS/eip-7906
Iván | ethrex
Reacted to "7906: https://eips.e..." with 🔥
Marc
our implementation gets data from the bal and logs being constructed
Iván | ethrex
Reacted to "7906: https://eips.e..." with 🚢
spencer
Reacted to "7906: https://eips..." with 🚢
Iván | ethrex
Replying to "https://faucet.frame..." Nethermind has been able to onboard today as validator
Alex Forshtat
https://ethereum-magicians.org/t/eip-7906-transaction-assertions-via-state-diff-opcode/23130/28?u=alex-forshtat-tbk
Iván | ethrex
Replying to "https://faucet.frame..." For other clients aiming the same, I’d recommend them to try to catch up with the chain and once you are confident enough on the implementation ask for some deposit tokens
Ben Adams
Could revert so not be in BAL if assertion fails
soispoke
https://ethresear.ch/t/mempool-account-transaction-capacity-from-historical-activity-matcha/25949
CPerezz
Reacted to "Could revert so not ..." with 👍
Guru
Reacted to "Could revert so not ..." with 👍
Iván | ethrex
Reacted to "https://ethresear.ch..." with 👀
Taek Lee
Reacted to "https://ethresear.ch/t/mempool-account-transaction-capacity-from-historical-activity-matcha/25949" with 👀
Iván | ethrex
Frames as Headliner on ACDE right?
spencer
Reacted to "Frames as Headline..." with ❤️
soispoke
A réagi à "Frames as Headline..." avec 🚢
Mega | Lambda
Reacted to "Frames as Headliner on ACDE right?" with ❤️
Mega | Lambda
Reacted to "Frames as Headliner on ACDE right?" with 🚢
Ben Adams
Reacted to "Frames as Headliner on ACDE right?" with 🚢
Ben Adams
Reacted to "Frames as Headliner on ACDE right?" with ❤️
Guru
Reacted to "Frames as Headliner ..." with 🚢

Call summary

Targets

•frames-devnet-0 — targeting launch this week, pending EthPandaOps availability - 00:06:17
•Frames Glamsterdam headliner ratification — Thursday ACDE call - 00:52:09

Decisions

•Toni recommends EIP-7906 (Transaction Assertions) be included in Glamsterdam; do not commingle assertion state diffs with BAL data - 00:48:21

Highlights

Client Updates:
- ·Geth, ethrex, and Nethermind all passing test suite and ready for frames-devnet-0; Nethermind joined ethrex public testnet as validator - 00:10:46
- ·Besu (Justin Florentine) has no code yet but is designating a point person; onboarding via ethrex testnet faucet - 00:45:12
Testing Progress:
- ·Test release v0.3.0 published; covers deep EVM code paths + frame-specific tests; sufficient for frames-devnet-0 readiness assessment - 00:08:07
Frame Expiry Discussion:
- ·raxhvl proposed replacing the expiry VERIFY frame with a native transaction-level 'features' field (key-value pairs, canonical ordering); saves ~32 bytes vs. empty frame encoding - 00:15:36
- ·lightclient and Toni skeptical: expiry is already implementable via VERIFY frame without state dependency; no evidence enough other features exist to justify generalizing; design discussion deferred to Ethereum Magicians / EthMAC - 00:24:42
Fork Status And Schedule:
- ·frames-devnet-0 blocked on EthPandaOps availability; Barnabas (out on vacation) is likely the facilitator; teams can use ethrex public testnet in the interim - 00:44:18
- ·Frames as Glamsterdam headliner EIP expected to be ratified at Thursday's ACDE call - 00:52:09
Eip Proposals Glamsterdam:
- ·EIP-7906 (Transaction Assertions via post-transaction frame + state-diff opcodes): Toni supports inclusion in Glamsterdam to avoid re-touching Frame Transactions later; recommendation to not commingle with BAL state diffs - 00:46:13
- ·ERC-1979 (key lifecycle standard for EIP-8141/EIP-8130): proposes common add/rotate/revoke key management with two tiers — recent-root (no state read, delayed revoke) vs. account state slot (instant revoke, one state read); cross-chain extension included - 00:50:41
- ·PR-12301 (maxVerifyGas as mempool floor): proposes redefining maxVerifyGas from a hard cap to a protocol-guaranteed floor, letting individual nodes accept heavier verify transactions above that floor - 00:41:01

Action Items

•raxhvl - raxhvl to post frame expiry / transaction features discussion to Frame Transaction EthMAC thread for async feedback - 00:29:05
•All client teams - Client teams to submit Hegota non-headliner EIP preferences to Nixo and Ansgar ≥24 hours before Thursday's ACDE - 00:52:17

Key decisions

Toni recommends EIP-7906 (Transaction Assertions) be included in Glamsterdam; do not commingle assertion state diffs with BAL data
IncludedRecommendation to include assertions to avoid re-touching Frame Transactions later, with a specific instruction to keep state diffs separate from BAL data
design discussion [for replacing expiry VERIFY frame with a native transaction-level features field] deferred to Ethereum Magicians / EthMAC
Proposal to save 32 bytes via a native features field was met with skepticism regarding the necessity of generalizing features
Geth, ethrex, and Nethermind all passing test suite and ready for frames-devnet-0
frames-devnet-0Client teams confirmed readiness following the publication of test release v0.3.0

EIPs discussed

EIP-7906

Links

AI Disclaimer: Some content or metadata on EIPsInsight may be AI-inferred or automatically compiled. If you find any discrepancy, please contact us at dev@avarch.org.