The EC byte-equality (corona_n1_byte_equality) proved the threshold combine
bit-equals CombineAbs, whose sign_abs_op_lifted_eq_rlwe bridge pins it to
rlwe_sign_op on the LAGRANGE-RECONSTRUCTED master secret -- exactly the path
the Boschini/Raccoon production combine must never take (CORONA-EC-RECON-MODEL).
This pass re-scopes that cone as IDEALISED CORRECTNESS and adds the HONEST
production model whose residual is a STANDARD Module-LWE/MSIS reduction.
New proofs/easycrypt/Corona_N1_NoLeak.ec (written; machine-recheck pending
EasyCrypt -- no `ec` on host, scripts/checks/ec-compile.sh is the CI gate):
- mask_telescope_zero [B, Lean-backed]: Sum_i (maskPrime_i - mask_i) = 0 --
the pairwise-PRF masks (z_i = R_i*u + maskPrime_i + c*lambda_i*s_i - mask_i)
are the same double sum reindexed (Fubini), so they cancel on aggregation.
Machine-checked in Lean as pairwise_mask_telescopes.
- no_leak_z_aggregate: the surviving aggregate is R*u + c*s by Lagrange-at-0
WITHOUT forming the master secret and with no per-party c*s_i ever exposed.
- no_leak_reduction [C-standard, OPEN]: under Module-LWE + Module-SIS the
public transcript leaks nothing about s beyond one Boschini signature. The
honest replacement for sign_abs_op_lifted_eq_rlwe -- a reduction to the SAME
substrate AXIOM-INVENTORY.md §1 already lists, secret NEVER reconstructed.
The CORRECTNESS CORE is machine-checked in Lean 4 + Mathlib on this host
(luxfi/proofs: Crypto.Corona.NoLeakAggregate / Threshold_Lagrange; lake build
green, #print axioms shows no sorryAx).
Disclosure updated to match reality: AXIOM-INVENTORY (new Bucket C-standard +
"A is machine-checked in Lean"), PROOF-CLAIMS (two models, §0 Disclosure 1),
BLOCKERS (CORONA-EC-RECON-MODEL re-scoped, first criterion now [x]), budget
AXIOM 56->58. Framework gate PASS (all 5 checks); go build + full test suite green.
reshare.Verify Lagrange-interpolates the input shares at X=0 and RETURNS the
reconstructed master secret to its caller (its own doc: "MUST NOT be used in
production"). Its only non-test caller is cmd/reshare_oracle (a KAT generator).
Left in the production package with no boundary, it is a latent master-key
footgun.
Fix (surgical split, not whole-file tag): reshare.go also defines the
production Reshare/Refresh/Share that keyera imports, so the file cannot be
tagged out. Move ONLY Verify into verify_reconstruct.go behind
//go:build corona_research; its shared interpolation helpers (selectQuorum,
lagrangeAtZero, sortedKeys) stay in reshare.go because production uses them.
Gate the sole non-test caller cmd/reshare_oracle/main.go with the same tag.
Tests: the reshare/refresh tests that reconstruct the secret (via Verify) move
into reshare_research_test.go / refresh_research_test.go / the tagged
integration_test.go and run under -tags corona_research. The production-
behaviour tests that do NOT reconstruct (determinism, arg validation, t=1
identity, fuzz, transcript) stay in the default build — no coverage lost.
Generic test helpers shared across both builds (cloneVector, runRoundsAndVerify)
move to untagged clone_test.go / signrounds_test.go.
Verification:
- default `go build ./...` exits 0; reshare.Verify ABSENT from the default
build (proven: a Verify-calling test does not compile into the default run).
- default `go test ./...`: 0 failures; production tests run.
- `go test -tags corona_research ./...`: green; reconstruction tests + oracle
compile and run (TestReshareSecretInvariant etc. pass).
- gofmt + vet clean in both build modes.
- security/framework/check.sh corona: PASS (no CLOSED-but-reconstructs blocker).
Reclassify (not reduce) the EasyCrypt axiom surface and correct the prior
inventory's understatement. NO axioms discharged this pass (honest): no
easycrypt toolchain on host, and every in-file-derivable axiom is already a
proved lemma; residual A/B axioms are over abstract ops/constants
(op group_pk_width : int., op sig_len : int.) and the C axioms are open.
- AXIOM-INVENTORY.md: replace the §2 table that mislabelled the wrapper
bridges and section-local module contracts as "Discharged in the
Wrapper". They are now correctly bucket C / OPEN: combine_body_spec hides
the Boschini combine steps 2-6, and combine_abs_op_lifted_bridge /
sign_abs_op_lifted_eq_rlwe pin the lifted op to rlwe_sign_op on the
Lagrange-RECONSTRUCTED master secret (reconstruct-then-sign). Complete
bucketed census of all 56 axioms -- A (18, 5 Lean-bridged), B (27), C
(11) -- one row per axiom with name, file:line, justification. CT
contracts (sign_round{1,2}_constant_time) classified C/fail-closed.
- PROOF-CLAIMS.md: add assurance vocabulary + two load-bearing disclosures
-- (1) the EC byte-equality MODELS reconstruct-then-sign; (2) Corona has
NO independent interop verifier (R-LWE has no NIST target), so its
byte-equality is asserted + same-construction Go<->C++ KAT, never
"proven/verified". Corrects the §4 claim that Pulsar is machine-checked
"at every step" (it too rests on the same reconstruct-then-sign cone).
- BLOCKERS.md (new): CORONA-EC-RECON-MODEL, CORONA-CT-PENDING,
CORONA-NO-INDEP-VERIFIER -- all OPEN, all cross-referenced to the shared
external-review gate.
- .assurance/reviewed-axioms.txt: list every load-bearing / security-cone
axiom (A: 5 Lean-bridged; C: combine_body_axiom, S_functional_spec,
combine_body_spec, combine_abs_op_lifted_bridge, sign_abs_op_lifted_eq_rlwe)
with bucket + justification.
- .assurance/budget.txt: correct prose (0 real admit. tactics; AXIOM=56 exact).
Framework gate (circular-proof + axiom-budget + the other three) PASS with
full disclosure. go test ./... green (no Go touched).
Closes the CRIT-1 residual (D1-2) sid-entropy gap and the CT/doc hygiene
gaps to bring Corona to the no-leak/no-gap bar. EasyCrypt proofs untouched.
PRIORITY 1 — anti-nonce-reuse / no-leak durability:
- The Round-1 nonce-PRF key no longer keys on a bare 64-bit sid. It now
derives from a 256-bit domain-separated SessionID (primitives.DeriveSessionID:
TranscriptHash("corona.sign.session-id.v1" || be64(sid) || T)) AND a fresh
per-signature 256-bit hedge salt drawn inside the kernel from party.Rand
(default crypto/rand). PRNGKeyForRound = PRF(skShare, "CoronaNonceV3" ||
sessionID || salt). Hedging restores threshold-Raccoon's fresh-per-signature
nonce posture: reuse durability no longer rests on the external consensus
layer never reissuing an sid — even an sid collision yields distinct R with
prob 2^-256. A deterministic 256-bit SessionID alone is necessary but NOT
sufficient (it repeats when sid repeats); the salt is what closes the leak.
- SignRound1 is fail-closed: rejects an all-zero derived SessionID or all-zero
salt with ErrDegenerateSession, and surfaces a short-read error. The
consensus slot-uniqueness invariant is documented as a HARD precondition at
SignRound1 and Signer.Round1 (no longer a buried comment).
- KAT/oracle determinism preserved via one seam: sign.DeterministicNonceSource
(KeyedPRNG over seed || "corona.sign.nonce-salt.v1" || partyIndex), set on
Party.Rand / Signer.SetNonceRand only by reproducibility harnesses.
- SignRound1 / Signer.Round1 now return an error; all call sites updated.
- KAT REGEN: only sign_verify_e2e.json changes (nonce-key bytes moved);
transcript_hash.json / MAC / legacy PRNGKey vectors are byte-stable, proving
the consensus-agreed transcript path was left untouched. Regenerated via
`bash scripts/regen-kats.sh`; `--verify` confirms byte-determinism (10 files).
- Regression: sign/nonce_reuse_test.go proves same-(skShare,sid) yields distinct
D (fresh R), pinned-nonce reproduces byte-identically, and the degenerate-
session guard fires. TestE2EKATReplayDeterminism rewritten to assert both the
hedged-differs and pinned-reproduces properties (was a defanged no-op).
PRIORITY 2 — constant-time hygiene:
- reshare/commit.go already used a constant-time comparator; the real gap was
the verbatim duplication of constTimePolyEqual+uint64SliceToBytes across dkg2
and reshare. Consolidated to one canonical utils.ConstantTimePolyEqual; both
delegate (no dkg2<->reshare dependency). Orphaned imports removed.
- FullRankCheck and the reshare commit path are now covered in the CT review
with their public-operand justification.
PRIORITY 3 — doc accuracy:
- CONSTANT-TIME-REVIEW.md rewritten Corona-specific and file:line-accurate:
drops the stale Pulsar/lens/warp/secp256k1 content; audits the real call
sites (hedged nonce key, masking PRF, lattigo samplers as the residual TCB
axiom, utils.ConstantTimePolyEqual, CheckL2Norm/Verify/FullRankCheck big.Int
variable-time on PUBLIC operands, activation/commit-digest array equality,
keyera/reshare zeroization).
- PROOF-CLAIMS.md §1: threshold.Combine / sign.LocalSign (nonexistent) -> the
real sign.Party.SignFinalize exposed as threshold.Signer.Finalize.
- threshold/threshold.go package doc: Ring-LWE -> Module-LWE.
Antje Worring