Evidence · reproduced live 27 Jul 2026

The evidence, reproduced live.

We don't ask you to trust a number on a slide. Every R-Series reference test below was re-run from a clean checkout of the frozen tag audit-tob-v1.1.0 on 27 Jul 2026 (Node v22.22.3). Here are the results, the exact commands, and — deliberately — what each one does not prove.

Five reports · reproduced, not asserted

Ground-demonstrated, not externally audited. The post-quantum leg uses real ML-DSA-65; key custody is a software signer, not a FIPS-certified HSM. Nothing here is flight-qualified. Every figure below traces to a run you can reproduce yourself with the command shown.

195
Reference suite
13 suites · 0 fail
15/15
Dual-anchor
permanence
150/150
Wycheproof
Ed25519
16/16
NIST ACVP
ML-DSA-65
33/33
R+12 Fabric
passport

Plus H-1 adversarial fuzz: 404,000 cases, 0 attacks succeeded. Frozen root: audit-tob-v1.1.0 · 697f339 · Node v22.22.3 · 0 production dependencies.

Mandatory non-claim (stated up front): Post-quantum signature: the current R-Series hybrid profile signs the post-quantum leg with real ML-DSA-65. Key custody: a software signer — hardware-backed HSM/KMS custody is a separate deployment control, is not deployed here, and no FIPS 140-3 / FIPS 204 / FIPS 205 validated module is claimed. Anchoring: the Base rail is live; the Bitcoin/OpenTimestamps rail is a deterministic stand-in and runs DARK by default — an anchoring rail, not the post-quantum signature leg. Not flight-qualified. Not independently audited by external third parties.

Reproduce every number below

git checkout audit-tob-v1.1.0   # frozen tag · commit 697f339
npm install                     # test deps only — 0 production deps
node test/run-all.mjs           # → ALL REFERENCE SUITES GREEN

Reference core: DCS-LabsAI/dcs-rseries-coreprivate repository; results reproduced internally from a clean checkout. Publicly reproducible artifacts: github.com/DCS-LabsAI/r2-standard. Each named suite below runs standalone too.


Report 1 · Reference implementation suite (LOG-03)

Continuous MMR transparency log + R2 receipt core

195 passed · 0 fail

What it proves. The R2 v1.1 reference suite passes in full: 14 suites, 404,208 checks executed, 0 failures, from a clean checkout. Composition, never conflated: 255 conventional assertions + 404,008 randomized/property checks, of which the H-1 adversarial suite contributes 404,000 ML-DSA cases with 0 successful attacks. Baseline: internal assurance candidate CORE-AUDIT-C2.9, root 391159c8…, 18 Aug 2026 (Gate A closed by independent internal verification; not an external audit). The append-and-prove path is correct with no counterexample found.

Result: 195 assertions, 13 suites, 0 failed (203 counting test_H2a_property.mjs's 8 property-assertions, reported as "ok"). Plus H-1 fuzz: 404,000 adversarial cases (malleability · downgrade · Merkle-forgery), 0 succeeded.

$ node test/run-all.mjs
✓ test.mjs 18 · keyreg 6 · interop 13 · groupA 24 · groupB 18 · groupC 18
✓ groupD 18 · groupE 20 · groupF 20 · groupG 16 · H2a ok · H2b 15 · H3a 9
ALL REFERENCE SUITES GREEN
Does not prove: that the AI action a receipt describes was correct or wise — only that the receipt is well-formed and unmodified.

Report 2 · Dual-anchor permanence

Base L2 + Bitcoin stand-in, single-chain-outage resilient

15 / 15 passed

What it proves. A root anchored to both rails verifies on each independently; when one chain is down the other still proves the root; when both are down the system reports UNANCHORED honestly rather than faking permanence. DARK by default — a live rail is refused unless its *_LIVE=1 flag is set.

Result: 15/15 — includes the negative cases (forged aggregate rejected, tampered inclusion path rejected, empty week rejected, hostile input never throws).

$ node --test test/dual_anchor.test.mjs
# tests 15 · pass 15 · fail 0
Does not prove: live on-chain permanence today. The Bitcoin rail is a deterministic stand-in and the Base anchor is a local stub labelled for the acceptance path; real chain writes require live network keys.

Report 3 · Wycheproof Ed25519 conformance

Google Project Wycheproof edge-case vectors

150 / 150 passed

What it proves. The verifier's accept/reject behaviour matches Google's public Wycheproof Ed25519 test vectors on all 150 edge cases — the adversarial suite the industry uses to catch signature-verification bugs.

$ node test/test_wycheproof_ed25519.mjs
Wycheproof Ed25519: 150 passed, 0 failed of 150
RESULT: PASS — verifier matches Wycheproof accept/reject on all edge cases
Does not prove: anything about the PQ leg — this covers classical Ed25519 only.

Report 4 · NIST ACVP ML-DSA-65 (post-quantum)

FIPS 204 accept/reject vectors — in-scope modes

16 / 16 in-scope · +44 skipped

What it proves. For the ML-DSA-65 modes the receipt layer actually uses, the verifier matches NIST's official ACVP accept/reject vectors, 16/16. The 44 skipped modes are ML-DSA parameter sets outside receipt-layer usage — out of scope, not failures.

$ node test/test_nist_acvp_mldsa65.mjs
NIST ACVP ML-DSA-65 sigVer: 16 passed, 0 failed, 44 skipped (modes outside receipt-layer usage)
RESULT: PASS — PQ leg matches official FIPS 204 accept/reject vectors
Does not prove: a FIPS-certified PQ implementation. The deployed signing path uses a software signer, not an HSM-certified signer; ACVP conformance here validates the verifier logic against NIST vectors, not a certified hardware signer.

Report 5 · R+12 Cross-Domain Provenance Fabric

Five domain roots → one verifiable passport

33 / 33 passed

What it proves. Software supply-chain, model lineage, AIBOM, compute, and content-disclosure roots bind into one MMR-anchored passport that a client verifies offline (7/7), with coverage computed and honestly enforced (absent domains commit a signed em-dash sentinel; over-claims are rejected two ways).

$ node test/fabric.test.mjs
R+12 FABRIC: 33 passed, 0 failed
[buyer/regulator verifies OFFLINE] PASSPORT VALID ✓ (7/7)
Does not prove: the domain roots are live production roots — the suite uses the pick-package test vectors.

What "reproduced" means here

Each figure on this page came from a command we ran on 27 Jul 2026 against the frozen tag, not copied from an earlier document. The repo is zero-production-dependency; the suites run with node --test, deterministic and clock-free. If a number here doesn't reproduce on your machine, tell us — we'll fix it or remove it. That's the deal.

Don't trust us. Reproduce it.

Clone the frozen tag, install test deps, run one command. Every number above resolves to a run you can execute yourself.