Agenda
Bi-weekly, Tuesdays @ 11:00 UTC [NEW TIME]
- Facilitator: Will Corcoran (will@ethereum.org · @corcoranwill)
- YouTube
- Previous Call
- Discord
- Spec PR
- Explainer
- Slides
- Prysm Write-up
Client Implementation Status
| Client | Status | Contact |
|---|---|---|
| Prysm | Old algo impl; 25% on latest spec | Terence |
| Lighthouse | EPF PoC; needs team review | — |
| Teku | PoC complete; production dev starting | Anton |
| Nimbus | Spec reviewed; active in spec PRs | Etan |
| Lodestar | Spec tests passing; deploying to staking env | Nazar |
| Grandine | TBD | — |
Testing
- Format: Extended fork choice tests (Altair+)
- Status:
is_one_confirmedtests merged; previous-epoch block confirmation tests in progress (~100+ tests, targeting ~95% coverage) - Paper: Finalizing; Roberto + Luca planning next steps
- Repo: [consensus-spec-tests TBD]
Call 04 Summary (Mar 3, 2026)
Headline
Spec approaching final state. Etan’s edge case fix landed with test coverage. New spec change improves FCR liveness without compromising security. Fork choice weight reuse explored and ruled out — FCR stays independent of fork choice, which is a net positive. No client devs on the call (ePBS devnet-0 likely pulling attention).
Client Updates
- No client developers present this call.
Spec Changes
| PR | What | Status |
|---|---|---|
| #25 | Etan’s edge case fix — first block in confirmation logic. Test written. Credits to Etan. | Merged |
| #28 | Adversarial threshold for honest FFG support — discount slashed validators from adversarial weight (same approach already used in is_one_confirmed). Increases liveness without compromising security. |
Open; some tests need fixing |
Fork choice weight reuse — decided against. After conversations with developers, reusing fork choice weights in FCR is not feasible as-is and would require additional tweaks. Upside of the decision: FCR implementation stays independent of fork choice, so the two components can be developed and changed in parallel.
Balance source clarified: FCR uses balances from the justified checkpoint at the beginning of the last slot of the previous epoch — aligned with the paper. Roberto confirmed the Lighthouse EPF implementation follows this approach.
Testing & Research
is_one_confirmedfunction tests (Luca) merged with good coverage.- Currently working on previous-epoch block confirmation tests — ~100+ tests targeting completion by mid-March. After done, estimated ~90-95% test coverage.
- Mikhail: once all tests pass for a client implementation, that gives meaningful assurance of correctness.
- Paper: Luca and Roberto meeting to plan final steps toward completion.
Action Items
| Owner | Item |
|---|---|
| Mikhail | Fix tests broken by #28; continue toward spec finalization |
| Luca | Complete previous-epoch block confirmation tests (~100+) by mid-March |
| Luca + Roberto | Finalize paper |
| All client devs | Review #28 (adversarial threshold change) |
| Will | Continue async outreach to client teams — attendance remains a priority |
Agenda for Call 05
A. Client updates (round-robin)
B. Spec status — #28 landed? Any remaining changes before spec freeze?
C. Testing — previous-epoch tests complete? Coverage assessment
D. Paper status
E. Lodestar staking env metrics (Nazar — carried from Call 04)
F. Open discussion
Add agenda items: Comment below with topic
Meeting Time: Tuesday, March 17, 2026 at 11:00 UTC (60 minutes)