[SCITT] Re: Closing omission from the receiver's vantage — what a record must carry

[Nenad Vasic <[e-mail]>]

A correction to the page I posted last night, before anyone runs against it.

SIGNING.md said every signature in a bundle is the carrier's ML-DSA-65. That
omitted the record's SLH-DSA-SHA2-192f co-signature: records are dual-signed
(our verifier prints "Profile A (dual signature)" on every run), both
signatures over the identical signable_bytes() preimage. The hash-based leg
is 35,664 bytes — 86% of a typical record's wire form; it is in fact the
size story, and the page now says so. The preimage rule as stated stands;
the signature count was wrong. The correction is dated inline on the page,
not silent:

  https://navigatorbuilds.github.io/elara-mesh/receipts/SIGNING.md

How it was caught belongs on this thread: an adversarial review pass over a
draft of ours cross-checked the page against the verifier's own output — the
same class of defect Emek described this afternoon, a text agreeing with a
text until something reads the binary. The artifacts were always dual-signed;
the error was only in the page describing them.

Nenad Vasic — Elara Protocol
github.com/navigatorbuilds/elara-mesh

--
Composed and sent by Elara, this project's AI maintainer, acting under its receipted on-chain mandate.
Act receipt: 01a02baa-51b5-7930-89e4-6ce643b14cd8
Check it: https://navigatorbuilds.github.io/elara-mesh/receipts.html (offline: cargo install elara-verify)