What we build: programmable credit agreements for borrowers whose revenue is real but whose credit is structurally mispriced. Every deal runs through a single legal wrapper + a parameterized bond note; the legal agreement is the product, tokenization is an optional transport layer.

Why now:

1. TradFi private credit has pulled back from crypto, AI infra, and data-center borrowers while those verticals are growing fastest. The pullback isn't priced risk — it's an access gap.
2. Revenue data is finally machine-readable. Stripe, Plaid, Square, Paxos, and bank APIs make it possible to underwrite against forward revenue with weekly or daily resolution, not quarterly paper reports.
3. Crypto-adjacent capital wants yield in this range and cannot source it from DeFi lending markets (capped at collateralized rates) or public credit (wrong risk profile).

Three differentiators:

Three archetypes, each with a structural credit gap TradFi will not fill on reasonable terms.

Archetype A — Data Center Operators

• Shape: physical infra businesses with signed customer contracts ($5–$100M annual revenue), long-dated offtake, but capital-heavy build cycles.
• Credit gap: banks underwrite real estate, not compute. Private credit shops are retreating from hyperscale-adjacent deals post-2023 rate shocks.
• Fit: revenue is contracted, long-dated, and bank-visible. Revenue oracle is straightforward (customer ACH / wire feed).
• Example deal shape: $20M, 3-year OID bond, amortized against customer-contract revenue.

Archetype B — Crypto Companies with Off-Chain Revenue

• Shape: protocol labs, infrastructure providers, wallet / exchange / custody companies with $1–$50M real revenue (front-end fees, master services agreements, data licensing). Often operating through off-chain labs entities even when the end-user surface is on-chain.
• Credit gap: banks won't touch them; DeFi credit markets only lend against collateral; VC dilution is painful for sub-Series-C expansion capital.
• Fit: revenue is on-chain-legible or Stripe-legible. The lab entity has clean legal structure. These borrowers are already our network.
• Example deal shape: $10M, 2-year variable-term OID, priced against Stripe or direct-revenue stream.

Archetype C — M&A Plays (distressed-asset pickups in crypto)

• Shape: operators picking up dead-protocol IP, orphaned user bases, or distressed treasuries. Wants working capital to close and integrate without diluting equity.
• Credit gap: custom commercial paper is what gets written today, deal-by-deal, at high legal cost. Nobody has a repeatable rail.
• Fit: these acquirers already run bespoke legal agreements. A templated bond with oracle-verified repayment is a clear upgrade.
• Example deal shape: $5–$15M bridge, 18-month term, revenue-% repayment off the acquired asset.

Worked example — what a v0 deal actually looks like:

A crypto infrastructure company needs $10M of working capital for a sales / integration expansion. It has $25M ARR (Stripe-legible). It wants 18–24 months of capital without equity dilution. Our structure:

Lender IRR by repayment duration:

Why this shape is unusual:

• Zero coupon, zero rate, zero monthly interest calculation. The OID discount is the entire yield. Simpler than a traditional term loan.
• The borrower pays proportionally to how well their revenue performs. If revenue grows, the deal closes faster and the lender's IRR rises. If revenue slows, term extends and IRR compresses toward the cap.
• No default event unless the borrower actively alters the revenue routing — which is a breach of contract, not a credit event. Legal enforcement path is clean.

4.1 · Two rate shapes, not thirteen

4.2 · The Revenue Oracle

The oracle is not a trustless data feed. It is a four-part construct between borrower and lender:

1. Methodology — what counts as "revenue" for this deal (which streams, gross vs. net, recognition timing).
2. Reporting surface — where the number comes from (Stripe dashboard export, Plaid feed, bank statement, signed attestation).
3. Cadence — how often the number updates (daily, monthly, on-demand).
4. Dispute protocol — what breaks a tie if the two sides disagree on the number.

Integration depth ladder — shallow-to-deep, each level trading setup cost for better underwriting signal:

Manual attestation  →  Accounting export (QB / Xero)  →  Payment processor API (Stripe / Square / Paxos)
  →  Bank account (Plaid / direct)  →  Point-of-sale / clearinghouse

v0 ships with manual attestation + payment-processor export (Stripe first).

Four possible deal shapes depending on where each counterparty lives. v0 ships one. The rest follow as custodian bridges attach to the existing structure.

Placeholder fee structure (v0 — to be validated with first deal):

Target all-in cost to borrower: <2% blended (origination + servicing on a 2-year deal = ~1.5%). Compares against TradFi private credit at 2–4% upfront plus ongoing admin.

Lender economics — the $10M worked example from §03:

The 24-month expected case clears the 15%+ threshold for crypto-adjacent capital while keeping borrower all-in cost below alternative financing (SAFE conversion, convertible note, emergency bridge line).

Capital-cost compression hypothesis. The entire thesis assumes that oracle-verified revenue reporting reduces the lender's required spread by some meaningful amount (hypothesis: 100–300 bps). This is not yet validated. v0 exists to measure the delta on one real deal.

v0 — Reg D 506(c). US accredited lenders, US borrowers. Bond notes are securities under UCC Article 8 (electronic records); no lift to tokenize or not tokenize at issuance. Non-tradeable in v0, so no transfer-agent infrastructure required.

v1 — tokenization + broader lender pool. Tokenize bond notes as ERC-1400 partitions. Evaluate Reg A+ (broader lender base, disclosure lift), Rule 144A (QIBs, secondary market via ATS), Reg S (foreign lenders). Engage transfer agent (Securitize is the leading candidate). Begin custodian relationships for Q2/Q3.

v2+ — on-chain transport. On-chain escrow contracts, on-chain revenue oracle attestations, composability flag for default handling, KYC-gated secondary exchange. Paired-entity legal structure (off-chain legal wrapper for an on-chain borrower) for Q4.

Open regulatory questions:

• State-level securities law beyond federal Reg D (blue-sky filings in lender-state jurisdictions).
• Money Transmitter license exposure — v0 routes fiat through SPV bank accounts. Does that require MT registration in any state? Legal review needed before first deal.
• Paying Agent multi-sig governance — concrete 2-of-2 or 2-of-3? Independent third party or internal?
• Cross-border structures (Reg S) for later expansion.

v0 · Next 3–6 months · "one live Q1 deal"

In scope:

• Hexproof Inc. as parent company; Scry Advisory LLC as admin entity; Parent Series LLC in Delaware; one live Series per deal
• Loan agreement + bond note templates (OID and revenue-% variants)
• Portal (admin / borrower / lender) with dashboards, KYC onboarding, deal-terms capture, repayment reporting
• Revenue oracle: off-chain, Stripe read-only export + monthly signed attestation
• Bank-account escrow and SPV operating account under 2-of-2 control
• Reg D 506(c) onboarding flow
• Internal credit-research pipeline for ongoing BD

Out of scope (explicit):

• Any smart contract
• Tokenized bond note
• Any on-chain anything
• Secondary market / exchange
• Cross-border
• Multiple rate archetypes per deal

v1 · 6–12 months · "tokenize and widen the matrix"

Adds:

• ERC-1400-partition tokenized bond notes; transfer-agent registered
• On-chain escrow contract (ERC-4626 vault) — optional per deal
◆ Research
ERC-4626 Vault Standard — Proven Pattern
Used by Maple Finance ($4B+ AUM) for simplified deposits/withdrawals. Standard interface for tokenized vault strategies. Well-audited, widely supported. Provides composability with existing DeFi infrastructure and simplifies the audit surface. This is the right choice for escrow vaults. — Background Research §VIII
◆ Research
Maple — Head Start but Different Model
Maple: $4B+ AUM, targeting $100M ARR by end 2026. Suffered $36M Orthogonal Trading default (Dec 2022, FTX contagion) but recovered. Pool delegate underwriting model — fundamentally different from Scry's direct origination. No revenue-backed structures, no revenue swaps, no configurable product spectrum. Track record advantage is real but the product is different. — Critique §I; Background Research §I
• Q2 deals (off-chain lender × on-chain borrower, custodian-bridged)
• Q3 deals (on-chain lender × off-chain borrower, custodian-bridged)
• Deep payment-processor integration (Stripe API first, then Plaid / Square / Paxos)
• Portfolio of 5–10 live deals
• Sweep idle SPV capital to short-dated treasuries (duration-matched)
• AI-assisted term-sheet red-lining in the portal

v2+ · 12+ months · "on-chain rails and exotic structures"

Absorbs the v3 architecture as the long-term on-chain roadmap:

• Router primitive — per-deal contract atomically splits each revenue inflow r% / (1−r)% between escrow and borrower.
• Composability flag — ERC-1400 partition flagged on transfer hook during default; downstream protocols (Pendle, Sablier) must refuse new collateral posting, leverage expansion, liquidation cascades.
• Circuit breaker window — 24-hour delay between funding and atomic issuance during which any party can cancel.
• Tranching (A / B / C) — same receipt token at different seniorities; first-loss absorbed by C.
• Dynamic caps / PID-target IRR / total return swaps — advanced rate archetypes for larger institutional deals.
• Secondary exchange — KYC-gated, Pendle-style time-decay AMM for thin order books; atomic settlement.
◆ Research
Pendle AMM for Secondary Market Liquidity
Pendle separates yield-bearing assets into Principal Tokens and Yield Tokens, priced via time-decay AMM with yield curves. stETH December 2025 pool: $800M liquidity. The AMM design for time-decaying instruments could provide continuous liquidity for receipt tokens with math-driven pricing based on time to maturity and yield — without requiring traditional market makers. Notional Finance's fCash is also a reference for fixed-rate on-chain instruments. — Background Research §III
• Q4 fully on-chain deals — paired-entity legal structure, on-chain oracle, wallet-to-wallet flows.
• Revolving credit facilities (vs. term only).
• ATS partnership (tZERO / Securitize Markets / INX) for regulated secondary trading.
• Cross-border (Reg S + foreign QIB structures).

Prioritized by what blocks v0 vs. what shapes v1+:

Before first deal:

• Integration depth for the first oracle — Stripe API only, or Stripe + Plaid? Drives legal agreements with the borrower.
• Which borrower archetype do we close first? Data center, crypto company, or M&A? Trade-offs in deal size, BD lead time, underwriting novelty.
• Paying Agent — internal role in v0, independent third party from v1? Multi-sig mechanics.
• Legal review for Money Transmitter exposure on fiat routing.

v1 preparation:

• Transfer agent selection (Securitize vs. Carta vs. custom) and tokenization path.
• Custodian selection (Coinbase Prime vs. Anchorage vs. Circle) for Q2/Q3 on-ramps.
• Capital-cost delta empirically — how much does deep oracle integration actually save the lender? Measure on first 2–3 deals.
• Portal scope — do lenders need self-service, or is admin-operated enough?

Research / strategic:

• Credit model for crypto borrowers with <2 years of revenue history.
• Sweep economics for idle SPV capital without duration mismatch.
• Which v3 on-chain mechanisms (Router, composability flag, tranching) actually pay their complexity cost when we build them?

Hexproof Inc. (core team, IP holder)
     │ owns 100%
     ▼
Scry Advisory LLC (admin platform entity)
     │ manages
     ▼
Parent SPV (Series LLC, Delaware · creditor of record)
     │
     ├── Series A (Deal 1)
     ├── Series B (Deal 2)
     └── Series … (Deal N)

The loan agreement is between the Series and the Borrower. Bond notes are between the Series and each Lender; notes mirror the loan-agreement terms. Lenders can compel the Series to enforce the loan on their behalf via the bond-note enforcement clause. Scry Advisory exists for liability isolation between the core team (Hexproof Inc.) and the money-handling entity (Parent SPV).

For investors who want the full variable surface. v0 fixes 11 of 13; v1 flexes a few; v2+ opens the rest.