LlamaRisk: Concluding Our Curve Engagement

Proposals
1

Summary

LlamaRisk will be concluding active engagement with Curve, effective June 30, 2026. The transition begins immediately. A detailed transition plan has been shared with Swiss Stake and key Curve stakeholders covering: completion of LlamaLend v1 deprecation work, resolution of impaired markets, initial LlamaLend v2 rollout preparation, emergency-control recommendations, and a full knowledge handoff to Swiss Stake and any successor risk service provider approved by the DAO.

We will cut our crvUSD payment stream after June 2026 and return the unused allocation to the DAO Treasury via governance payload. The CRV stream cannot be cut early without burning the unvested portion; LlamaRisk will retain no compensation after June 30, and all CRV vesting beyond that date will be returned to the DAO treasury.

A successor risk service provider is a decision for Curve DAO. We will support a clean and complete handoff to whichever path the DAO chooses.

A personal note. Wormhole, CEO

Curve has been formative to LlamaRisk. The idea for the service began during my time on the Curve Grant Council, where I worked on identifying ecosystem talent and supporting their work. The initial concept - an independent risk provider conducting deep due diligence on pegged assets and harnessing analyst talent from within the Curve community - would not have left the page without early belief from Curve’s leadership.

Michael Egorov was the first active advisor to our team. His support was instrumental in helping LlamaRisk gain the credibility we needed to secure funding from the Grants Council and, later, from Curve DAO directly. Curve became our first client in 2022 and has been our longest continuous engagement ever since. Everything LlamaRisk has grown into as a firm, including the trust we have built with other DeFi protocols, traces back to that foundation.

We owe a great deal to the Curve community, and this post is, more than anything, an attempt to wind that engagement down with the same care and seriousness that defined it.

What is changing

LlamaRisk has been awarded an expanded mandate at Aave as its sole external risk service provider. Honoring that commitment requires us to phase out our other engagements, including the one with Curve. This is a structural decision about where LlamaRisk concentrates its resources going forward, not a reflection of our view of Curve or of the work we have built together.

We renewed with Curve in April 2026 with the intent to serve through April 2027. The change in our broader engagement structure has made that timeline no longer feasible. The responsible course of action is to be direct, set a clear end date, and ensure the work we leave behind is in the best possible shape.

Transition end date and scope

Active engagement ends June 30, 2026. Between now and that date, our work is focused on five workstreams:

  1. LlamaLend v1 deprecation — executing the work outlined in the v1 market deprecation plan, including monetary policy changes, borrow/loan discount adjustments, gauge actions, and governance payload preparation.
  2. Impaired-market resolution — advancing remediation for the bad-debt markets and sDOLA. Active proposals are already on the forum: sDOLA long2 repayment and CRV long LlamaLend market recovery. Our objective is full implementation before departure; where that is not feasible, we will leave a documented, stakeholder-reviewed resolution path.
  3. Initial LlamaLend v2 rollout preparation — recommended parameters, onboarding framework, monitoring framework, and coordination with Resupply for an initial set of markets on Optimism and Ethereum mainnet, pending v2 deployment timelines. If rollout timing slips, we will still deliver the frameworks and readiness materials.
  4. Emergency controls and owner-proxy recommendations — written recommendations to expand eDAO emergency authorities relevant to crvUSD stability and LlamaLend v2, to ensure rapid response capability after our departure.
  5. Knowledge transfer and handoff — full handoff package for Swiss Stake and any successor team, including a transition memo, risk framework summary, outstanding governance items list, repo and document access, status appendices for v1 deprecation / impaired markets / v2 onboarding, and direct onboarding sessions if a successor is in place before our exit.

Beyond these workstreams, our scope through June 30 is limited to continuity of pre-existing recurring obligations. New market onboardings, integrations, and net-new analytical workstreams are out of scope unless explicitly agreed in writing.

Funding treatment

We want to be unambiguous about the stream termination plan.

crvUSD stream. We will prepare a governance payload to cut the crvUSD stream effective after June 2026. The payload includes a clawback of the unvested allocation back to the DAO Treasury.

CRV stream. The CRV stream cannot be cut early without burning the unvested portion. LlamaRisk will retain no compensation after June 30, 2026 — all CRV vesting beyond that date will be returned to the DAO treasury within reasonable timelines as token vest. Swiss Stake will be notified when repayments are made.

The intent is that the DAO pays for what it receives, and nothing more.

Successor coverage

The decision on a successor risk service provider sits with Curve DAO. We will support a complete handoff to whichever team the DAO approves, including direct onboarding sessions and transfer of the conceptual frameworks and operating considerations behind our work. If no successor is in place at the time of our exit, the full handoff package will be delivered to Swiss Stake so the work can continue without operational discontinuity. What we can commit to is leaving the work in a state where a successor, whoever that ends up being, can pick it up cleanly.

Thank you

To Michael Egorov: thank you for believing in this idea before there was anything to believe in, and for the active support that gave LlamaRisk its first foothold. The work LlamaRisk has been able to do over the past four-plus years exists in large part because of that early conviction.

To the Swiss Stake team: working alongside you has been a privilege. Your development and research teams have sharpened ours at every turn, and the community managers’ diligence and care for the protocol have been a model of how a core team can hold an ecosystem together. Curve is fortunate to have you, and we are grateful for the partnership.

To the broader Curve community: the stakeholders, contributors, integrators, and users who have engaged with our research and pushed our thinking; thank you. We will do our utmost to leave the work in a state worthy of the trust you have placed in us.

The team remains fully present and engaged through June 30. We will continue posting transition progress updates through our regular channels and welcome questions from the community in this thread.

— Wormhole CEO, LlamaRisk

Disclaimer
The selection of any successor risk service provider and any due diligence on a candidate’s qualifications or fitness rests entirely with Curve DAO and the broader Curve community. LlamaRisk neither endorses nor recommends any specific candidate and assumes no responsibility for the successor’s selection, performance, conduct, or any decisions, omissions, or actions the successor takes following our exit. After June 30, 2026, LlamaRisk will have no continuing role, duty, or obligation in respect of Curve, LlamaLend, crvUSD, or any related market and will not monitor parameters, respond to incidents, propose governance actions, or maintain emergency authorities. The handoff materials are provided as of their respective publication dates, reflect LlamaRisk’s research and operational view at that time, and carry no representations or warranties — express or implied — as to fitness for purpose, future protocol performance, market outcomes, or continued accuracy after the publication date; any party relying on these materials after our departure does so at its own risk and should re-validate them against then-current conditions. Nothing in this post or in any handoff material constitutes legal, tax, or investment advice, and the post should not be read as a continuing commitment by LlamaRisk to any party beyond the active engagement period.

2 Likes
2

Thank you for the clear notice, and for committing to leave a documented, stakeholder-reviewed resolution path rather than simply closing the door. That commitment is the right call — and it’s exactly what this question is about.

For the impaired markets in workstream #2 that won’t be fully resolved by June 30, the single most useful thing the handoff can do is be explicit about one distinction: which markets are being handed over with a funding-and-liquidity path, and which are being handed over with a monitoring framework only. Those are very different things to inherit, and a successor needs to know which is which on day one.

CRV-long is the clearest example of why it matters. After about a month with a live gauge, the crvUSD side of the recovery pool still sits around 211,600 crvUSD against a cvcrvUSD side in the hundreds of millions, and the pool quotes the claim near 0.00081. The mechanism is sound, but it never reached the depth where holders can actually exit — so as of today this is a market with a recovery design but no working exit.

Two ways to close that gap are already on the forum record, and neither asks LlamaRisk to author anything new before leaving:

The ask is small: name these as open items in the handoff — even a single line flagging that the depth gap is unfinished and pointing the successor at the two candidates. If the intended handoff for this market really is monitoring-only, that’s an understandable scope call given the timeline; it simply helps everyone to have it stated plainly, so the gap is owned rather than left implicit.

Either way — thank you for the four years.

1 Like