ONyc — Risk Report
High risk · 4.0/10
What this actually is
ONyc is OnRe’s tokenized reinsurance yield coin on Solana — a proportional claim on a Bermuda-regulated segregated account that underwrites short-duration insurance and reinsurance contracts. The reported yield, 11.64% APY (OnRe, July 2026 review), comes from real reinsurance premium income plus collateral yield on an sUSDe float. That makes the return source genuinely less correlated to crypto than most DeFi yield products.
⚠️ The catch is equally important: ONyc holders are the first-loss insurance capital. If catastrophe claims or underwriting losses exceed premiums and reserves, NAV falls directly. There is no senior tranche sitting above holders. The book is roughly one year old and has not been tested through a major loss year.
Disambiguation. “Onyx” is spelled ONyc and reads like “onyx.” The OnRe sibling token ONe is a legacy structured-LP product and is not covered here — if you hold something labelled “ONe,” confirm its current status directly with OnRe.
⚠️ A scope note that governs every figure below: ONyc is Solana-only and this coverage does not currently run an independent read of it. Every number here is OnRe’s own disclosure, not a measurement of ours. That is a gap in what we monitor for this asset specifically, not an absence of Solana-reading capability — our tooling reads Solana for other assets.
1 · Stability — 5.0
Reference: NAV. ONyc is not a stablecoin and should not be evaluated as a fixed $1 peg. It is a NAV-tracking vault share, and NAV has moved from about $1.00 to about $1.13, with the token recently trading close to it.
⚠️ That smooth history reflects a young, calm period rather than a tested loss cycle. Holders are first-loss underwriting capital: the mechanism by which NAV falls is a claims year, and there has not been one yet. A flat line on a one-year-old first-loss instrument is an absence of evidence, not evidence of stability.
⚠️ NAV is operator-posted, which is a further reason not to read the smoothness as independently confirmed.
2 · Backing — 4.0
What stands behind a share is a reinsurance book plus a collateral float, held inside a Bermuda Segregated Accounts Company. OnRe cites Class IIGB insurance and Class F digital-asset business licensing, Coinbase Prime custody, and top-tier reinsurance brokers including Guy Carpenter and Howden. That institutional wrapper is real and it is the strongest thing about this asset.
⚠️ What is missing is verification. NAV is operator-posted, the reserves are not independently attested here, and capital managed on-chain — $245.81M (OnRe, 2026-07-31) — is a disclosure rather than a read. A regulated wrapper constrains behaviour; it does not substitute for a look-through.
⚠️ One figure from the same source is deliberately omitted. The July review’s “$357.21M across OnRe DeFi Markets” against $245.81M managed on-chain cannot be reconciled from public disclosure — gross-of-recycled, cumulative-deployed and notional readings are all consistent with it — and a number that large reads as scale when it may be a measure of leverage. It will not appear here until it can be explained.
3 · Liquidity & Exit — 3.5
Both exit paths, and here both are constrained. The axis takes the worse one, and they are close.
Primary — gated, and bounded by something you can watch. KYC and accredited-investor status required, about 2.5% of NAV per period into a pending queue, settling in USDC or USDG when liquidity allows, filled at the prevailing NAV at fulfilment — so if NAV moves while you are pending, you bear it. The protocol does not force-unwind live reinsurance contracts to meet withdrawals.
⚠️ That last point is what makes capital utilisation the binding constraint. OnRe reports utilisation at 94.65% (2026-07-31) — roughly 5% of the book is uncommitted capital, and redemptions are served from that free slice. The 2.5% gate is not an arbitrary throttle; it is calibrated to sit inside a thin slice.
In normal conditions this works — 5% free against a 2.5% period draw is roughly two periods of cover, and the queue clears. ⚠️ Under stress the arithmetic tightens from both ends at once: redemption demand rises exactly when utilisation is highest, and new premium income is the main thing that replenishes the free slice. A rise in utilisation is an early-warning signal on exit, and it moves before the queue visibly slows — and unlike the queue, OnRe publishes it.
Secondary — thin. Roughly $2.7M across Kamino, Orca, Raydium and Loopscale against $245.81M managed on-chain — about 1.1% of the book — with no centralized-exchange listing.
⚠️ This is severe exit asymmetry, and under stress the primary queue and secondary depth can tighten at the same time. Treat ONyc as an illiquid, long-hold credit / insurance position, not a cash-equivalent yield token.
4 · Dependencies — 4.5
⚠️ Two risks are stacked here, and the stacking is the finding — neither one is the whole story.
1. Reinsurance underwriting — the core exposure and the source of the yield. Premiums are earned up front, but large claims hit NAV, and holders absorb them first. Reinsurance is a large, established market and the exposure is genuinely uncorrelated to crypto. That is the good half.
2. ⚠️ The collateral float is sUSDe, so Ethena sits underneath the insurance book. Idle and reserve capital earns through sUSDe, which brings Ethena’s basis-trade, funding and peg risks beneath everything above. A holder is exposed to a bad claims year and to a funding-rate regime change, and those two have nothing to do with each other — which means diversification between them, but two independent ways to lose rather than one.
Add the operational dependency: the whole position runs through a single Solana program and OnRe’s own marking of NAV.
5 · Contract & Admin — 4.0
⚠️ On-chain control is single-key, and this is the sharpest gap between the legal wrapper and the code.
The addresses, so this is checkable rather than asserted — re-measured 2026-09-11, with a fabricated pubkey returning null in the same pass:
| role | account | what it is |
|---|---|---|
| ONyc mint | 5Y8NV33Vv7WbnLfq3zBcKSdYPrk7g2KoiQoe7M2tcxp5 | SPL Token mint, 9 decimals |
| program upgrade authority | FvmhydbpHGQzMUp51GmhB1fwsrkyfmnRsTg7oPwDe25f | ⚠️ System-owned, 0 bytes — a plain wallet |
| freeze authority | 45YnzauhsBM8CpUz96Djf8UG5vqq2Dua62wuW9H3jaJ5 | ⚠️ System-owned, 0 bytes — a plain wallet |
| mint authority | AbpE5YLpdpxj2jRczG9P341Jicf67NvZsaZYrATbMnNX | ✅ owned by OnRe’s program — a PDA |
| OnRe program | onreuGhHHgVzMWSkj2oQDLDtvvGvoepBPkqyaubFcwe | BPF upgradeable loader |
✅ The mint is identified by the pointer back rather than by name: its mint authority is a PDA owned by OnRe’s own program, which is what ties this mint to this issuer. ⚠️ Both privileged wallets are System-Program-owned with zero bytes of account data — that is the shape of an ordinary keypair. An SPL Token multisig would be Token-program-owned; a Squads vault would be program-owned. Neither is. ⚠️ As everywhere on this site, a bare account proves there is no ON-CHAIN quorum and says nothing about off-chain MPC or HSM custody, which cannot be read from a chain. One key can redeploy the mint and redemption logic, or freeze holder tokens.
The token is a classic Solana SPL mint, 9 decimals. ⚠️ Supply re-measured 2026-09-11: 270,454,463.60 ONyc — up from 266,271,281.72 two days earlier, and up 26.6% from the 213,548,263 read on 2026-07-22. The sleeve has grown by roughly a quarter in seven weeks and is still growing, so any figure on this page dated to July describes a materially smaller book than the one that exists now. Minting is mediated by an OnRe program PDA, but the program itself is upgradeable by that single wallet. A Quantstamp audit exists for the OnRe Solana implementation.
⚠️ The Bermuda regulated wrapper is a real legal backstop, but it is off-chain. It does not remove the on-chain key risk, and an audit describes the code as written rather than who may replace it.
6 · Issuer — 5.0
OnRe, a Bermuda Segregated Accounts Company. ⚠️ The institutional wrapper and partners are meaningful positives — Class IIGB and Class F licensing, Coinbase Prime custody, Guy Carpenter and Howden as brokers, and a segregated-account structure that legally ring-fences the book.
⚠️ What holds this axis at 5.0 rather than higher is that the on-chain reality does not match the off-chain one. Single-key upgrade and freeze authority, operator-posted NAV, and a disclosure set that this coverage cannot independently verify. Institutional legal wrapper, centralized on-chain trust — the two halves point in opposite directions and the score sits between them.
Who should avoid
- Anyone who needs redeemable-at-par liquidity, a permissionless primary exit, or deep secondary markets. The practical retail exit is thin DEX liquidity unless you can satisfy KYC and accreditation (axis 3).
- Anyone who requires multisig or timelock governance before taking token risk. Upgrade and freeze are single-key (axis 5).
- Anyone who does not want first-loss insurance exposure, or stacked Ethena exposure (axis 4).
- Anyone reading the yield as a savings rate. It is compensation for catastrophe risk, sUSDe collateral risk, a redemption gate and single-key trust (axes 1 and 4).
ONyc is best sized, if at all, as a small long-hold sleeve for investors who explicitly want reinsurance-premium exposure and can tolerate delayed or impaired exit.
What to watch
- Capital utilisation. 94.65% at 2026-07-31, and it moves before the queue visibly slows — the single best early warning on exit (axis 3).
- Whether upgrade and freeze authority ever move to a multisig or timelock. The largest available improvement (axis 5).
- A first real claims year. The book is untested, and that is what the yield is paid for (axes 1 and 4).
- Whether the $357.21M figure is ever reconciled against the $245.81M managed on-chain (axis 2).
- Ethena’s funding regime, which sits underneath the collateral float (axis 4).
This report is built from publicly available documentation, third-party market data, and independent Solana RPC reads dated 2026-07-22. We hold no privileged information about OnRe, its contracts, or its insurance book. Corrections welcome to info@tidresearch.com.
Revision history: 2026-09-09 — supply re-measured on-chain and it has grown 24.7% in seven weeks, 213,548,263 → 266,271,281.72 ONyc. ⚠️ market_cap_approx was ABSENT and is now populated at approximately $301M — the asset had been rendering with no size at all. Its basis is deliberately mixed and stated as such: supply is measured, NAV ~1.13 is operator-posted and carried from July rather than re-verified. ✅ The two single-wallet authority findings were re-checked and both still hold — freeze authority and program upgrade authority are each a bare account with no on-chain quorum. No scores changed: nothing measured here contradicts the existing axes, and a 24.7% growth in a sleeve whose risks are structural rather than size-driven is not by itself a re-score. 2026-08-23 — capital utilization added; figures attributed and dated; all scores held. The substantive gap was that this report explained the redemption gate without ever naming the constraint behind it — the word “utilization” did not appear anywhere in it. OnRe’s July 2026 review reports ONyc utilization at 94.65% as of 2026-07-31, so roughly 5% of the book is uncommitted capital, and that free slice is what redemptions are paid from given the report’s own statement that OnRe does not force-unwind live reinsurance contracts. The 2.5%-per-period queue is calibrated to sit inside it. No axis moved: Redemption 3.5 already priced a gated, capacity-limited exit, and this is the missing evidence for that score rather than a reason to change it. Figures refreshed and attributed: reported yield stated as 11.64% (OnRe, July 2026 review) rather than “around 11%”, which rounded the wrong way; capital managed on-chain stated as $245.81M (OnRe, 2026-07-31), replacing “roughly $240M” in two places. Two figures were checked and deliberately left alone: NAV “about $1.13” is accurate against $1.1294, and the roughly 213.5M supply is explicitly dated to the July verification and remains correct as a dated fact. One figure from the same source is deliberately omitted: the review’s “$357.21M across OnRe DeFi Markets” against $245.81M managed on-chain cannot be reconciled from public disclosure — gross-of-recycled, cumulative-deployed and notional readings are all consistent with it — and a number that large reads as scale when it may be a measure of leverage. It will not appear here until it can be explained. ONyc is Solana-only and this coverage does not currently run an independent read of it, so every figure in this report is OnRe’s disclosure rather than a measurement of ours; last_verified is not bumped, since nothing here was independently verified. To be precise about why: this is a gap in what we monitor for ONyc specifically, not an absence of Solana-reading capability — our tooling does read Solana for other assets. 2026-07-23 — initial retail publish: new ONyc report; overall 4.0 with six-axis vault-share scoring; first-loss reinsurance exposure, sUSDe collateral stacking, single-key upgrade/freeze control, KYC/accredited redemption gate, and thin secondary liquidity are the binding risks.