Paper research · advisory · not live

Proof of Reserves — honestly

A proof of the methodology, not of real reserves. We show that NAV reconciles from its parts and is cryptographically tamper-evident — on a simulated book.

Read this first

This is PAPER. $0 real capital. There is no custody, no real reserves, and no client funds. The book is a simulated $100,000 USDC portfolio. The figures below prove that NAV reconciles correctly from its components and is tamper-evident — not that real reserves exist.

real capital / custody

$100k

simulated book (USDC)

methodology

is what's proven — not reserves

NAV reconciliation (paper book)

source: static / offline

Every (simulated) dollar accounted for, cryptographically verifiable

The NAV is recomputed from its parts (deployed + cash + accrued) and reconciled against the reported equity. This is a paper proof-of-reserves: the verifiable methodology on the virtual $100k book, not real custody.

loading…

Deployed

—

Cash

—

Accrued

—

NAV (= equity)

—

Reconciliation delta

—

Positions

—

Last verified

—

Verification hash (NAV)

verifiable · tamper-evident

—

components: —

NAV = Σ(per-protocol positions) + cash + accrued yield. components_hash = sha256 of the canonical JSON of the sorted positions; nav_hash binds the NAV to its components and timestamp. Anyone can recompute the published NAV from its components. All amounts are simulated (paper).

What this is NOT

Not proof of real on-chain reserves or a custodial balance.
Not a claim that client funds or external capital are under management.
Not a Merkle-proof attestation over user deposits (there are no users and no deposits).
It is proof that our own accounting reconciles and is tamper-evident — the same discipline a live book would need, rehearsed on paper.

Verify it yourself

The desk's proof-bearing decision chains and exit schedules verify under the zero-dependency standalone verifier — with none of our code. The same "don't trust, check" discipline applies across every public artifact.

python3 verify_spa.py data/rates_desk/ python3 verify_spa.py data/rates_desk/ --replay python3 verify_spa.py data/dd_snapshot/ --offline

The --replay flag goes beyond hashes: it re-derives each verdict from its own published numbers (fair_yield = baseline − total haircut; approved ⇒ net_edge > 0; every refusal carries a reason). A stored verdict that doesn't follow from its inputs fails closed — proving not "we didn't edit the row" but "the decision actually follows from the numbers". --offline checksums a frozen, self-contained DD snapshot against its pinned manifest — a funder replays the whole dataset with no live API and no trust in us.

verifier: scripts/verify_spa.py · spec: docs/PROOF_CHAIN_SPEC.md

Track Record

Evidenced days, not padded

Refusal log

What we declined, and why

Liquidation-NAV

What you realize on exit