Risk assessment
C
Partner vaults only
MAY 2026
frxUSD/scrvUSD LP as collateral in the frxUSD v2 vault
The underlying assets are sound. The pool is not deep enough to liquidate against.
Overall grade C, driven entirely by the market layer.
What we assessed
The collateral is the Stake DAO LP token for the Curve frxUSD/scrvUSD pool, proposed for the
Stake DAO frxUSD v2 vault. Because frxUSD is the loan asset of that vault, only
scrvUSD was carried through the asset layer.
Layer results
| Layer | Driver | Score | Grade |
|---|
| Asset rating — scrvUSD | Operational pillar | 3 | B |
| Platform rating — Curve | Standing grade | 5 | S |
| Market rating | Pool health | 2 | C |
Why the market layer set the grade
The pool clears the trading-frequency and pool-age tests comfortably, but two criteria pull
it to 2: pool TVL sits near the bottom of the grid, and LP concentration is severe, with the
largest provider holding a large minority of the pool excluding protocol-owned liquidity.
- Underlying asset liquidity scores at the top of the grid: the cap is small relative to total available depth
- Price fluctuation scores mid-grid on max drawdown against the proposed LLTV
- Pool TVL and LP concentration are the binding constraints
Could not verify
The price feed sub-pillar was left unscored pending confirmation of the oracle
configuration for the proposed market. The grade below is therefore provisional on that point.
Recommendation
Eligible for a partner vault mandate only, at a conservative absolute cap
rather than a share-of-supply cap, with a re-assessment triggered on any material change in
pool depth or LP concentration.
Risk assessmentC
2026
pmUSD partner vault: a mandate built around a single issuer
Placeholder entry. Content to be written from the existing assessment file.
This slot demonstrates how a second assessment sits in the reader. Each publication carries
the same structure: what was assessed, the layer results, the binding constraint, what we
could not verify, and the resulting recommendation.
Asset ratingB
2026
scrvUSD: yield-bearing crvUSD and the v3.0.3 audit gap
Strong reserves and external ratings, held back by an operational pillar that scores on the
weakest of its criteria.
Where the grade comes from
The asset rating takes the worst pillar, and each pillar takes the worst criterion within it.
For scrvUSD the operational pillar is the binding one: the audit criterion scores low because
the Yearn V3 version actually deployed was not covered by the audits that cover the earlier
releases, and no other curator currently uses the asset as collateral, which caps the peer
review criterion.
Verified onchain
Contract immutability and reserve visibility were confirmed against contract state, not taken from the issuer questionnaire.
Asset ratingA
2026
sfrxUSD: reserve composition and redemption path
Placeholder entry. Content to be written from the existing research file.
The reader pane keeps a fixed measure of about seventy characters so long assessments stay
readable, with tables and callouts breaking up the argument.
Methodology note2026
How we size caps: from observable liquidity to liquidation ability
A cap is a claim about how much collateral can be sold, in a stressed market, before the
position goes underwater. Here is how we turn that into a number.
Methodology notes explain the framework rather than grade an asset. They are the documents
a prospective client reads to decide whether our reasoning is one they can rely on.