<?xml version='1.0' encoding='utf-8'?>
<!DOCTYPE rfc [
  <!ENTITY nbsp    "&#160;">
  <!ENTITY zwsp   "&#8203;">
  <!ENTITY nbhy   "&#8209;">
  <!ENTITY wj     "&#8288;">
]>
<?xml-stylesheet type="text/xsl" href="rfc2629.xslt" ?>
<!-- generated by https://github.com/cabo/kramdown-rfc version 1.7.43 (Ruby 3.2.3) -->
<?rfc rfcedstyle="yes"?>
<rfc xmlns:xi="http://www.w3.org/2001/XInclude" ipr="trust200902" docName="draft-hillier-chorale-protocol-00" category="exp" submissionType="IETF" tocInclude="true" sortRefs="true" symRefs="true" version="3">
  <!-- xml2rfc v2v3 conversion 3.34.1 -->
  <front>
    <title abbrev="Chorale Protocol">The Chorale Protocol: Interception-Resistant Packet Transmission with Topology-as-Secret Ordering, Cascade-Integrity Witnessing, Polyglot Cover Packets, Per-Packet Time-Lock Sealing, and Multi-Jurisdiction Substrate Diversity</title>
    <seriesInfo name="Internet-Draft" value="draft-hillier-chorale-protocol-00"/>
    <author initials="J. D." surname="Hillier" fullname="Joel David Hillier">
      <organization abbrev="Certisyn">Certisyn, Inc.</organization>
      <address>
        <postal>
          <country>US</country>
        </postal>
        <email>jhillier@certisyn.com</email>
      </address>
    </author>
    <date year="2026" month="September" day="27"/>
    <area>Security</area>
    <keyword>confidentiality</keyword>
    <keyword>post-quantum</keyword>
    <keyword>one-time pad</keyword>
    <keyword>traffic analysis</keyword>
    <keyword>cover traffic</keyword>
    <keyword>verifiable delay function</keyword>
    <keyword>multi-substrate</keyword>
    <keyword>defence</keyword>
    <keyword>finance</keyword>
    <keyword>sovereign</keyword>
    <keyword>hash agility</keyword>
    <abstract>
      <?line 169?>

<t>This document specifies the Chorale Protocol, a secure packet
transmission system with active integrity assurance, for use cases
requiring resistance to interception, traffic analysis, replay,
tampering, and retrospective decryption by adversaries with future
quantum capability. Chorale defines three confidentiality regimes,
computational, everlasting and information-theoretic, and states the
keying conditions under which each may be relied upon. The
information-theoretic regime requires an independently corroborated
co-presence or quantum-key-distribution strand consumed as an
unexpanded pad. A session keyed by a derived keystream is
computational and is reported as computational. A conforming
implementation fails closed and refuses to report a regime its keying
does not support. The protocol composes a per-session secret graph topology
that determines packet ordering without transmitting any ordering
information on the wire, cascade-integrity witnessing over the hidden
topology, cover packets carrying valid-looking witnesses for fictional
sessions, per-packet Verifiable Delay Function time-lock sealing,
multi-source physical-entropy one-time pad composition with sovereign
jurisdictional separation, multi-substrate flight requiring threshold
reconstruction across diverse path types, and a tombstone ledger that
makes pad reuse structurally impossible at the receiver. This version
introduces hash-family negotiation at handshake, raising the default
hash function for HKDF key derivation, VDF time-lock sealing, and
server-blind relay commitments from SHA-256 to SHA-512, with SHA3-512
admitted as a Keccak-family alternative for adversary-evolution-
tolerant postures and SHA-256 retained as a backward-compatible legacy
mode for resource-constrained devices. The protocol is intended for
use by intelligence agencies, central banks, treaty-bound corridors,
regulator-to-regulator communications, and other settings demanding
the strongest practical confidentiality and integrity properties.</t>
    </abstract>
  </front>
  <middle>
    <?line 202?>

<section anchor="introduction">
      <name>Introduction</name>
      <t>Existing transport-layer cryptographic protocols including <xref target="RFC9846"/>
provide confidentiality based on computational hardness assumptions
that are not robust to large-scale fault-tolerant quantum computation.
Post-quantum key encapsulation mechanisms (see <xref target="FIPS203"/>) mitigate
the long-term risk but do not eliminate the threat of harvest-now-
decrypt-later collection of ciphertext. Information-theoretic
confidentiality via one-time pads <xref target="Shannon1949"/> provides provable
secrecy independent of adversary computational power, but practical
deployment is constrained by the key-distribution problem and by
operational failures arising from pad re-use.</t>
      <t>This document specifies the Chorale Protocol, a packet-based secure
transmission system that composes pad-based confidentiality with a
set of mechanisms that defend the pad-based scheme against the
operational and physical-layer attacks that have historically broken
one-time pad deployments in practice. The confidentiality regime a
session attains is determined by how its keying material reaches the
payload, and is specified in <xref target="regime-floor"/>. The protocol is designed for
use by parties able to perform a one-time co-presence event to
establish session anchor material, including intelligence agencies
operating within allied corridors, central banks operating within
multilateral settlement frameworks, regulator-to-regulator filings
within mutual-recognition treaty zones, and other contexts in which
the additional setup cost is warranted by the confidentiality
requirements.</t>
      <section anchor="conventions-and-terminology">
        <name>Conventions and Terminology</name>
        <t>The key words "<bcp14>MUST</bcp14>", "<bcp14>MUST NOT</bcp14>", "<bcp14>REQUIRED</bcp14>", "<bcp14>SHALL</bcp14>", "<bcp14>SHALL
NOT</bcp14>", "<bcp14>SHOULD</bcp14>", "<bcp14>SHOULD NOT</bcp14>", "<bcp14>RECOMMENDED</bcp14>", "<bcp14>NOT RECOMMENDED</bcp14>",
"<bcp14>MAY</bcp14>", and "<bcp14>OPTIONAL</bcp14>" in this document are to be interpreted as
described in BCP 14 <xref target="RFC2119"/> <xref target="RFC8174"/> when, and only when, they
appear in all capitals, as shown here.</t>
        <?line -18?>

<t>This document uses the following terms:</t>
        <dl>
          <dt>Session Anchor:</dt>
          <dd>
            <t>A per-session secret comprising at least a topology graph, a
one-time pad, a Merkle root over expected plaintexts, a duress
codeword, and a post-quantum digital signature binding the anchor
to the sender's verifiable credential. Established during a
co-presence event.</t>
          </dd>
          <dt>Topology Graph:</dt>
          <dd>
            <t>A graph G = (V, E) with N nodes wherein each node v is associated
with a set N(v) of at most K named neighbour-nodes. The edge
structure is a per-session secret never transmitted over any
network substrate.</t>
          </dd>
          <dt>Atomic Packet:</dt>
          <dd>
            <t>A single unit of message decomposition corresponding to exactly one
node of the topology graph.</t>
          </dd>
          <dt>Neighbour-Witness:</dt>
          <dd>
            <t>A cryptographic hash, attached to each atomic packet, of the
payloads of the K neighbours of the corresponding node in the
topology graph.</t>
          </dd>
          <dt>Cover Packet:</dt>
          <dd>
            <t>A packet structurally indistinguishable from an atomic packet but
carrying random payload and a neighbour-witness computed against a
fictional session topology generated fresh per cover-packet batch.</t>
          </dd>
          <dt>Tombstone Ledger:</dt>
          <dd>
            <t>A synchronised append-only record at both sender and receiver
endpoints of per-pad-section consumption.</t>
          </dd>
          <dt>Substrate:</dt>
          <dd>
            <t>A network channel suitable for packet transport. Multiple
heterogeneous substrates (terrestrial fibre, satellite, microwave,
mesh radio, sneakernet courier, sovereign dedicated fibre) are used
in combination.</t>
          </dd>
          <dt>Co-presence Event:</dt>
          <dd>
            <t>A one-time event during which the sender and receiver establish the
Session Anchor via a channel guaranteeing secrecy against any
non-endpoint adversary.</t>
          </dd>
          <dt>Hash Family:</dt>
          <dd>
            <t>One of a small set of cryptographic hash function families
permitted for use as the underlying primitive of HKDF key
derivation <xref target="RFC5869"/>, VDF iterative sealing, and server-blind
relay commitments within a session. Negotiated once per session at
handshake and bound for the lifetime of the session.</t>
          </dd>
        </dl>
      </section>
    </section>
    <section anchor="protocol-overview">
      <name>Protocol Overview</name>
      <t>The Chorale Protocol comprises the following steps:</t>
      <ol spacing="normal" type="1"><li>
          <t><strong>Co-presence</strong>. The sender and receiver establish a Session
Anchor via a co-presence event <xref target="co-presence"/>. The anchor <bcp14>MUST</bcp14>
include a Topology Graph and a one-time pad. The anchor <bcp14>SHOULD</bcp14>
include a duress codeword, a Merkle root over expected
plaintexts, and a post-quantum digital signature <xref target="FIPS204"/>
binding the anchor to the sender's verifiable credential.</t>
        </li>
        <li>
          <t><strong>Hash-family negotiation</strong>. As part of the rendezvous payload
exchanged at handshake, the sender and receiver each advertise a
list of supported hash families. They converge on the strongest
mutually-supported family per the algorithm in <xref target="hash-family"/>.
The negotiated family is bound for the lifetime of the session
and drives all subsequent HKDF, VDF, and commitment-layer
computations.</t>
        </li>
        <li>
          <t><strong>Pad generation</strong>. The one-time pad is generated by combining,
via bitwise exclusive-or, the outputs of at least two and
preferably at least three uncorrelated physical entropy sources
resident in distinct sovereign jurisdictions <xref target="multi-sovereign"/>.</t>
        </li>
        <li>
          <t><strong>Decomposition</strong>. The sender computes ciphertext C as the
bitwise exclusive-or of the plaintext message with a portion of
the one-time pad. The sender records the pad consumption in the
Tombstone Ledger. The sender optionally applies a per-packet
reading-frame offset, then partitions C into N Atomic Packets
each corresponding to one node of the Topology Graph.</t>
        </li>
        <li>
          <t><strong>Witnessing</strong>. For each Atomic Packet P_i corresponding to node
v_i, the sender computes a Neighbour-Witness W_i = H(P_{N(v_i)[1]}
|| P_{N(v_i)[2]} || ... || P_{N(v_i)[K]}) where H is the hash
function selected by the negotiated hash family (see
<xref target="hash-family"/> and <xref target="crypto-profile"/>). The Neighbour-Witness is
appended to the packet.</t>
        </li>
        <li>
          <t><strong>Time-lock sealing</strong>. For each Atomic Packet, the sender computes
a Verifiable Delay Function time-lock seal S_i = VDF(P_i || W_i,
T) requiring T sequential operations to compute and to invert.
The seal is appended to the packet. See <xref target="vdf"/>.</t>
        </li>
        <li>
          <t><strong>Cover packet generation</strong>. The sender generates a quantity of
Cover Packets at least equal to the quantity of Atomic Packets.
Each Cover Packet has random payload and a Neighbour-Witness
computed against a fictional session topology generated fresh per
cover-packet batch. The fictional topology <bcp14>MUST</bcp14> be discarded
after dispatch.</t>
        </li>
        <li>
          <t><strong>Flight</strong>. The sender transmits Atomic Packets and Cover Packets
over two or more Substrates of heterogeneous type. The release
schedule is modulated by prime-modular timing <xref target="flight"/>.
Multi-substrate flight <bcp14>MUST</bcp14> be configured such that no single
substrate carries more than K-1 of the M shares required for
reconstruction, where K and M are system parameters.</t>
        </li>
        <li>
          <t><strong>Reassembly</strong>. The receiver collects transmitted packets across
an arbitrary time window. For each packet, the receiver attempts
to match the Neighbour-Witness against the receiver's local copy
of the Topology Graph. Packets whose witnesses do not validate
are classified as Cover Packets and discarded. For each Atomic
Packet, the receiver validates the Neighbour-Witness against the
K-neighbour-packet payloads. If any validation fails, the
receiver enters a cascade-failure state and aborts reassembly.</t>
        </li>
        <li>
          <t><strong>Time-lock inverse</strong>. For each Atomic Packet, the receiver
computes the inverse of the time-lock seal S_i in parallel
across multiple physical processors.</t>
        </li>
        <li>
          <t><strong>Plaintext reconstruction</strong>. The receiver orders the validated,
unsealed packet payloads according to the Topology Graph,
producing the ciphertext stream C. The receiver applies the
inverse of the per-packet reading-frame offset, if applicable.
The receiver computes the plaintext P = C XOR K[i:i+L].</t>
        </li>
        <li>
          <t><strong>Final verification</strong>. The receiver computes the Merkle root of
P and compares to the Merkle root shared at co-presence. If the
roots match, the plaintext is accepted; otherwise the session
is rejected.</t>
        </li>
        <li>
          <t><strong>Tombstone update</strong>. Both endpoints update the Tombstone Ledger
to mark consumed pad sections. Any subsequent attempt to consume
a marked section <bcp14>MUST</bcp14> be structurally rejected at the receiver.</t>
        </li>
      </ol>
    </section>
    <section anchor="co-presence">
      <name>Co-presence Event</name>
      <t>The Session Anchor <bcp14>MUST</bcp14> be established via a co-presence event
through one of the following channels:</t>
      <ul spacing="normal">
        <li>
          <t>In-person meeting of the sender and receiver, with the anchor
written to identical removable media at each endpoint;</t>
        </li>
        <li>
          <t>Quantum Key Distribution <xref target="BB84"/> session between endpoints;</t>
        </li>
        <li>
          <t>A sovereign trusted exchange operating under bilateral or
multilateral treaty (e.g., the AUKUS dedicated communications
corridor);</t>
        </li>
        <li>
          <t>Any other channel for which the parties can demonstrate that the
anchor cannot be observed by any non-endpoint adversary.</t>
        </li>
      </ul>
      <t>The Session Anchor includes:</t>
      <ul spacing="normal">
        <li>
          <t>The Topology Graph G with N nodes and K-degree neighbour
structure (<bcp14>RECOMMENDED</bcp14> N &gt;= 64, K = 7, minimum girth &gt;= 5);</t>
        </li>
        <li>
          <t>The one-time pad K of length sufficient for the projected message
volume of the session;</t>
        </li>
        <li>
          <t>A session identifier known only to the endpoints;</t>
        </li>
        <li>
          <t>A Merkle root computed over one or more expected plaintexts;</t>
        </li>
        <li>
          <t>A duress codeword known only to the endpoints;</t>
        </li>
        <li>
          <t>A post-quantum digital signature <xref target="FIPS204"/> binding the anchor
to a verifiable credential of the sender.</t>
        </li>
      </ul>
      <t>The Topology Graph and the one-time pad <bcp14>MUST NOT</bcp14> be transmitted over
any network substrate after the co-presence event has completed.</t>
    </section>
    <section anchor="hash-family">
      <name>Hash Family Negotiation</name>
      <t>Implementations <bcp14>MUST</bcp14> select a single hash family per session at the
rendezvous-and-handshake step, and <bcp14>MUST</bcp14> bind that selection for the
lifetime of the session. This section defines the registered
families, the negotiation algorithm, and the binding semantics.</t>
      <section anchor="registered-hash-families">
        <name>Registered Hash Families</name>
        <t>The following hash families are defined by this document.</t>
        <table>
          <thead>
            <tr>
              <th align="left">Identifier</th>
              <th align="left">Underlying primitive</th>
              <th align="left">Default for</th>
            </tr>
          </thead>
          <tbody>
            <tr>
              <td align="left">sha-256</td>
              <td align="left">SHA-256 <xref target="FIPS180-4"/></td>
              <td align="left">Backward-compatible legacy / constrained dev</td>
            </tr>
            <tr>
              <td align="left">sha-512</td>
              <td align="left">SHA-512 <xref target="FIPS180-4"/></td>
              <td align="left">Sovereign and Allied-corridor deployments</td>
            </tr>
            <tr>
              <td align="left">sha3-512</td>
              <td align="left">SHA3-512 <xref target="FIPS202"/></td>
              <td align="left">Adversary-evolution-tolerant posture</td>
            </tr>
          </tbody>
        </table>
        <t>The <tt>sha-256</tt> identifier is RETAINED as the legacy backward-
compatible family. Resource-constrained devices that cannot
accommodate the 64-byte block size of SHA-512 <bcp14>MAY</bcp14> advertise
<tt>sha-256</tt> only. Implementations supporting <tt>sha-256</tt> <bcp14>MUST NOT</bcp14> use
shorter digests for HKDF or VDF computations.</t>
        <t>The <tt>sha-512</tt> identifier is the <bcp14>RECOMMENDED</bcp14> default for new
deployments. It is the mandatory-to-implement family for sovereign
corridor and Allied-corridor profiles, and is the negotiation default
when both peers advertise it.</t>
        <t>The <tt>sha3-512</tt> identifier is <bcp14>OPTIONAL</bcp14>. It is <bcp14>RECOMMENDED</bcp14> for
adversary-evolution-tolerant postures where deployment is intended
to remain resilient against future cryptanalytic advances in the
Merkle-Damgard SHA-2 construction. The Keccak sponge construction is
structurally independent of the Merkle-Damgard lineage, providing
cryptographic-construction diversity against the same threat class
that the multi-source entropy generator addresses at the entropy
layer.</t>
      </section>
      <section anchor="rendezvous-payload-field">
        <name>Rendezvous Payload Field</name>
        <t>The rendezvous payload exchanged at handshake <bcp14>MUST</bcp14> include a
<tt>hash_family</tt> field carrying the list of hash family identifiers the
sending peer supports, ordered by descending preference. The field
<bcp14>MUST</bcp14> be a JSON array of strings drawn from the registered set above.</t>
        <t>Example rendezvous payload fragment:</t>
        <artwork><![CDATA[
{
  "session_id": "...",
  "ecdh_public_jwk": { ... },
  "hash_family": ["sha3-512", "sha-512", "sha-256"]
}
]]></artwork>
      </section>
      <section anchor="negotiation-algorithm">
        <name>Negotiation Algorithm</name>
        <t>On receipt of the peer's rendezvous payload, each endpoint <bcp14>MUST</bcp14>
compute the negotiated hash family as follows:</t>
        <ol spacing="normal" type="1"><li>
            <t>Form the intersection of the local supported list and the remote
advertised list.</t>
          </li>
          <li>
            <t>If the intersection is empty, the session <bcp14>MUST</bcp14> be aborted with a
<tt>no-common-hash-family</tt> error. Implementations <bcp14>MUST NOT</bcp14> silently
downgrade to an out-of-set algorithm.</t>
          </li>
          <li>
            <t>From the intersection, select the strongest family by the
following preference ordering, from strongest to weakest:  </t>
            <artwork><![CDATA[
sha3-512  >  sha-512  >  sha-256
]]></artwork>
          </li>
          <li>
            <t>Both endpoints, executing the same algorithm against the same two
advertised lists, <bcp14>MUST</bcp14> reach the same selection. Disagreement is
a protocol violation and <bcp14>MUST</bcp14> abort the session.</t>
          </li>
        </ol>
        <t>The negotiated identifier <bcp14>MUST</bcp14> be carried in the Session Anchor and
<bcp14>MUST</bcp14> be recorded in the session record on both endpoints for audit.</t>
      </section>
      <section anchor="binding-semantics">
        <name>Binding Semantics</name>
        <t>Once negotiated, the hash family is bound for the lifetime of the
session. The negotiated family <bcp14>MUST</bcp14> drive:</t>
        <ul spacing="normal">
          <li>
            <t>HKDF <xref target="RFC5869"/> key derivation of the session topology graph
bytes and the one-time pad bytes from the ECDH shared secret;</t>
          </li>
          <li>
            <t>The HMAC <xref target="FIPS198-1"/> primitive used for any per-session MAC key
derivation and per-packet packet-authentication tag;</t>
          </li>
          <li>
            <t>The VDF iterative-hash construction used for per-packet time-lock
sealing (see <xref target="vdf"/>);</t>
          </li>
          <li>
            <t>The server-blind relay commitment digest length and computation
(see <xref target="commitment"/>).</t>
          </li>
        </ul>
        <t>The negotiated hash family <bcp14>MUST NOT</bcp14> drive the digital-signature
hash used by the ECDSA-P256 signature primitive of <xref target="FIPS186-5"/>,
which remains SHA-256 per the canonical pairing of the curve. The
hash-family negotiation surface of this section is explicitly
scoped to HKDF, VDF, MAC, and commitment computation. Signature
hash agility is out of scope for this version.</t>
        <t>A session <bcp14>MUST NOT</bcp14> change hash families mid-flight. A peer wishing
to upgrade the family for subsequent traffic <bcp14>MUST</bcp14> establish a new
session via a fresh co-presence event.</t>
      </section>
      <section anchor="hash-family-sas">
        <name>SAS Binding of the Negotiated Family</name>
        <t>The out-of-band Short Authentication String (SAS) recited at the
end of the rendezvous-and-handshake step <bcp14>MUST</bcp14> cover the negotiated
<tt>hash_family</tt> byte in addition to the ECDH public keys. The SAS
preimage is constructed as:</t>
        <artwork><![CDATA[
SAS_preimage = hash_family || "|" || SAS_words
]]></artwork>
        <t>and the SAS digest is computed under the negotiated family:</t>
        <artwork><![CDATA[
SAS_digest = H_negotiated( SAS_preimage )
]]></artwork>
        <t>Implementations <bcp14>MUST</bcp14> also include the <tt>hash_family</tt> identifier in
the HKDF <tt>info</tt> label that derives the SAS keystream so the SAS
words themselves diverge across families even before the digest
step. Either binding alone would suffice; both together provide
defence in depth.</t>
        <t>This binding is what makes an active man-in-the-middle attempting a
hash-family downgrade detectable at the SAS step. See
<xref target="sec-mitm-downgrade"/> for the threat model and analysis.</t>
      </section>
    </section>
    <section anchor="multi-sovereign">
      <name>Multi-Sovereign Entropy Generator</name>
      <t>This section governs the generation of pad material for the
<tt>everlasting</tt> and <tt>information-theoretic</tt> regimes of <xref target="regime-floor"/>.
Implementations <bcp14>MUST</bcp14> generate that pad material by combining outputs
of at least two uncorrelated physical entropy sources by bitwise
exclusive-or. Implementations <bcp14>SHOULD</bcp14> use at least three sources
resident in at least three distinct sovereign jurisdictions to ensure
that compromise of any single source (whether by adversary infiltration,
supply-chain attack, or sovereign-government coercion) does not reduce
the effective entropy of the pad.</t>
      <t>Recommended source types include:</t>
      <ul spacing="normal">
        <li>
          <t>A DNA sequencer producing noise from per-base call-quality jitter
of an ongoing biological sequencing run;</t>
        </li>
        <li>
          <t>A quantum random number generator producing noise from a vacuum-
fluctuation or radioactive-decay source;</t>
        </li>
        <li>
          <t>An atmospheric radio noise generator producing noise from a wide-
band radio receiver tuned to an unallocated frequency band.</t>
        </li>
      </ul>
    </section>
    <section anchor="hkdf">
      <name>HKDF Key Derivation</name>
      <t>Implementations using the client-keyed handoff (the standard form
for new deployments) derive the session Topology Graph bytes and the
payload keystream bytes from an ECDH shared secret via HKDF
<xref target="RFC5869"/>.</t>
      <t>A keystream derived in this way is not a one-time pad. Expansion of a
shared secret cannot increase its entropy, and the shared secret is
itself established under a computational assumption. A session keyed
by this path therefore attains the <tt>computational</tt> regime of
<xref target="regime-floor"/> and <bcp14>MUST</bcp14> report it as such. Implementations <bcp14>MUST NOT</bcp14>
describe a derived keystream as a one-time pad, and <bcp14>MUST NOT</bcp14> claim a
raised regime on the strength of the pad-generation requirements of
<xref target="multi-sovereign"/> unless the generated material reaches the payload
unexpanded.</t>
      <t>The HKDF hash function <bcp14>MUST</bcp14> be the SHA-2 or SHA-3 function
corresponding to the negotiated hash family of <xref target="hash-family"/>:</t>
      <ul spacing="normal">
        <li>
          <t><tt>sha-256</tt>   =&gt;  HKDF-SHA-256</t>
        </li>
        <li>
          <t><tt>sha-512</tt>   =&gt;  HKDF-SHA-512  (<bcp14>RECOMMENDED</bcp14> default)</t>
        </li>
        <li>
          <t><tt>sha3-512</tt>  =&gt;  HKDF-SHA3-512</t>
        </li>
      </ul>
      <t>HKDF emits an arbitrary-length output keystream and the on-wire
materialisation of derived bytes is unchanged across families. The
substantive change introduced by raising the default from
HKDF-SHA-256 to HKDF-SHA-512 is in the internal mixing margin: the
larger internal state and block size of SHA-512 give a wider
collision-resistance band against any future cryptanalytic
discovery affecting the SHA-2 family's compression function.
HKDF-SHA3-512 admits a Keccak-construction alternative for
deployments wishing to diversify away from the Merkle-Damgard
construction altogether.</t>
      <t>Implementations <bcp14>MUST</bcp14> use distinct <tt>info</tt> strings for the topology
and pad derivations so that the same shared secret produces
unrelated keystreams. The <bcp14>RECOMMENDED</bcp14> <tt>info</tt> values are
<tt>chorale-topology-N{N}-K{K}</tt> for the topology and
<tt>chorale-keystream-{padBytes}</tt> for the payload keystream.</t>
    </section>
    <section anchor="vdf">
      <name>Verifiable Delay Function Sealing</name>
      <t>Implementations <bcp14>MUST</bcp14> attach a Verifiable Delay Function (VDF) seal
to each Atomic Packet. Two constructions are admitted:</t>
      <ul spacing="normal">
        <li>
          <t>An algebraic VDF in an RSA group, per <xref target="Wesolowski2018"/> and
<xref target="BBBF2018"/>, for deployments holding the hardware to evaluate
modular squarings at production rate;</t>
        </li>
        <li>
          <t>An iterative-hash VDF based on the negotiated hash family, for
deployments preferring a software-only path with no RSA group
dependency.</t>
        </li>
      </ul>
      <t>The iterative-hash VDF baseline is 200 rounds of iterated
application of the negotiated hash family to the concatenation of
the packet payload and the neighbour-witness, with each round's
output fed as the input to the next. Where the negotiated family is
<tt>sha-256</tt> the construction is equivalent to the v-01 baseline.
Where the negotiated family is <tt>sha-512</tt> or <tt>sha3-512</tt> the per-round
work approximately doubles relative to the SHA-256 baseline, since
SHA-512 and SHA3-512 each process the same number of input bits per
round through a larger internal state. Both sender and receiver pay
this work symmetrically; the cost is bounded at protocol setup by
the choice of round count and hash family.</t>
      <t>The 200-round count is a deployment baseline. Implementations
<bcp14>MAY</bcp14> raise the round count for stronger sequential-cost bounds at
the corresponding increase in send and verify latency. The round
count <bcp14>MUST</bcp14> be carried in the seal field so the receiver can
reproduce the construction without negotiating round count
separately.</t>
      <t>The rationale for the doubled per-round work is that the VDF is
the anti-replay and anti-shortcut primitive against future
adversaries who may develop parallel-execution shortcuts on
SHA-2-family iteration. A larger per-round state surface raises
the bar for any such shortcut. The cost trade is symmetric: the
sender's send-time cost rises in lockstep with the adversary's
inversion cost, so an honest deployment is not disadvantaged by
the upgrade.</t>
    </section>
    <section anchor="commitment">
      <name>Server-Blind Relay Commitment Layer</name>
      <t>The Chorale relay layer is server-blind: relays carry opaque bytes
and a fixed-length commitment to per-session material (topology
hash, pad hash, short-authentication-string hash) without holding
any preimage. Commitment integrity at the relay defends against an
adversary who compromises the relay infrastructure and attempts to
substitute material at the rendezvous step.</t>
      <t>The commitment digest length <bcp14>MUST</bcp14> match the negotiated hash family
of <xref target="hash-family"/>:</t>
      <ul spacing="normal">
        <li>
          <t><tt>sha-256</tt>   =&gt;  32-byte SHA-256 commitment (legacy)</t>
        </li>
        <li>
          <t><tt>sha-512</tt>   =&gt;  64-byte SHA-512 commitment (<bcp14>RECOMMENDED</bcp14> default)</t>
        </li>
        <li>
          <t><tt>sha3-512</tt>  =&gt;  64-byte SHA3-512 commitment</t>
        </li>
      </ul>
      <t>The upgrade from 32-byte SHA-256 to 64-byte SHA-512 commitments
doubles the collision-resistance band of the commitment digest
while increasing per-session storage at the relay by 32 bytes per
commitment (negligible for any practical deployment). The 32-byte
SHA-256 commitment <bcp14>MUST</bcp14> be retained as a legacy mode for
backward-compatible negotiation with constrained-device endpoints
advertising only <tt>sha-256</tt>.</t>
      <t>The relay <bcp14>MUST NOT</bcp14> interpret commitment bytes. The relay <bcp14>MUST</bcp14>
record the digest length advertised in the session record and <bcp14>MUST</bcp14>
reject mismatched-length commitment-update attempts within a single
session.</t>
    </section>
    <section anchor="crypto-profile">
      <name>Cryptographic Profile</name>
      <t>Implementations <bcp14>MUST</bcp14> support SHA-256 <xref target="FIPS180-4"/>, SHA-512
<xref target="FIPS180-4"/>, and SHA3-512 <xref target="FIPS202"/> for Neighbour-Witness
computation, with the choice of function within a session bound by
the negotiated hash family of <xref target="hash-family"/>. Implementations
<bcp14>SHOULD</bcp14> also support SHA3-256, BLAKE2b, and BLAKE3 for tenant-
selected digest profiles outside the negotiated set.</t>
      <t>Implementations <bcp14>MUST</bcp14> support Ed25519 <xref target="FIPS186-5"/> and ECDSA-P256
<xref target="FIPS186-5"/> for digital signatures. ECDSA-P256 signatures are
canonically paired with SHA-256 per <xref target="FIPS186-5"/>; this pairing is
NOT subject to the hash-family negotiation of <xref target="hash-family"/>.
Implementations <bcp14>SHOULD</bcp14> also support ECDSA-P384 <xref target="FIPS186-5"/> for
Commercial National Security Algorithm Suite 2.0 transitional
compatibility.</t>
      <t>Implementations <bcp14>MUST</bcp14> support CRYSTALS-Dilithium / ML-DSA <xref target="FIPS204"/>
for the post-quantum signature on the Session Anchor.</t>
      <t>The selection of cryptographic profile per session is a tenant
configuration parameter. The selection <bcp14>MUST</bcp14> be recorded in an
auditable activation log such that the algorithm used for any past
envelope can be determined.</t>
    </section>
    <section anchor="flight">
      <name>Packet Flight</name>
      <t>Atomic Packets and Cover Packets are released into the network with
release schedule modulated by per-session prime-modular timing.
Specifically, each packet is released at a time t_i congruent to a_i
(mod p) where p is a per-session-secret prime number and a_i is a
per-packet release offset derived from session-derived randomness.</t>
      <t>Multi-substrate flight is configured such that no single substrate
carries more than K-1 of the M shares required for reconstruction.
<bcp14>RECOMMENDED</bcp14> parameters: K = 5, M = 8.</t>
    </section>
    <section anchor="tombstone-ledger">
      <name>Tombstone Ledger</name>
      <t>Each pad section is assigned a unique tombstone identifier at
issuance. On consumption, the tombstone is marked spent. Any
subsequent attempt to consume a spent tombstone <bcp14>MUST</bcp14> be structurally
rejected at the receiver endpoint, regardless of whether the sender
intentionally or accidentally re-uses a pad section.</t>
      <t>The Tombstone Ledger <bcp14>MUST</bcp14> be maintained synchronously at both sender
and receiver endpoints.</t>
    </section>
    <section anchor="duress-codeword">
      <name>Duress Codeword</name>
      <t>If the Session Anchor includes a duress codeword, use of the duress
codeword in place of the authentic session codeword in a packet
payload <bcp14>MUST</bcp14> cause the protocol to complete successfully at the
receiver  -  to avoid alerting the adversary that the duress has been
detected  -  while silently flagging the receiver-side audit log to a
duress condition. Subsequent packets received after the duress flag
is set <bcp14>SHOULD</bcp14> be treated as potentially compromised and routed to a
duress-quarantine queue.</t>
    </section>
    <section anchor="cascade-integrity-property">
      <name>Cascade Integrity Property</name>
      <t>The Chorale Protocol provides cascade integrity over the hidden
topology. If an adversary captures the dispatched packets and
modifies a single bit of any single packet, that modification
invalidates the Neighbour-Witness recorded for that packet at each
of the K packets that name it as a neighbour. Because the adversary
does not know the Topology Graph, the adversary cannot identify
which K packets must be additionally modified to maintain consistency.
Each attempted single-packet forgery cascades to invalidate K+1
Neighbour-Witnesses at the receiver.</t>
      <t>For Topology Graphs of girth &gt;= 5 and K = 7, the probability that an
adversary modifying random subsets of packets can maintain Neighbour-
Witness consistency without knowledge of the topology decreases
exponentially with the number of packets modified.</t>
    </section>
    <section anchor="polyglot">
      <name>Polyglot Cover-Encoding Layer</name>
      <t>Cover packets, as defined in <xref target="flight"/>, are by themselves cryptographically
indistinguishable from real packets at the witness layer. However, an
adversary performing traffic-shape analysis on a single substrate may
still derive useful information from the <em>encoding presentation</em> of the
captured stream: if every packet on the wire is base64-encoded ciphertext
of similar length, the substrate itself becomes a fingerprint indicating
that it carries Chorale-like traffic.</t>
      <t>This section defines an <bcp14>OPTIONAL</bcp14> extension which generates each cover
packet in one of a heterogeneous set of presentation encodings, with the
encoding assignment derived from a session-bound seed shared between
sender and receiver.</t>
      <section anchor="cover-encoding-set">
        <name>Cover Encoding Set</name>
        <t>Implementations supporting this extension <bcp14>MUST</bcp14> be able to generate cover
packets in each of the following presentation encodings:</t>
        <ul spacing="normal">
          <li>
            <t><tt>base64-binary</tt>   -  random bytes encoded as base64 (the legacy form)</t>
          </li>
          <li>
            <t><tt>hex</tt>             -  random bytes encoded as lowercase hexadecimal</t>
          </li>
          <li>
            <t><tt>english-prose</tt>   -  syntactically valid English text from a wordlist</t>
          </li>
          <li>
            <t><tt>german-prose</tt>    -  syntactically valid German text from a wordlist</t>
          </li>
          <li>
            <t><tt>mandarin</tt>        -  Mandarin Chinese filler phrases</t>
          </li>
          <li>
            <t><tt>arabic</tt>          -  Arabic filler phrases</t>
          </li>
          <li>
            <t><tt>json-noise</tt>      -  JSON-shaped key-value pairs with random values</t>
          </li>
          <li>
            <t><tt>log-line</tt>        -  structured log line shape (timestamp + level + msg)</t>
          </li>
        </ul>
        <t>Real packets <bcp14>MUST</bcp14> continue to use the <tt>base64-binary</tt> presentation
because the receiver decode path requires deterministic byte recovery.
Only cover packets carry the polyglot presentation.</t>
      </section>
      <section anchor="per-packet-encoding-selection">
        <name>Per-Packet Encoding Selection</name>
        <t>For each cover packet at index <tt>i</tt> in a batch of <tt>N</tt> cover packets, the
sender selects the encoding as:</t>
        <artwork><![CDATA[
idx         = H(session_id || cover_batch_nonce || i) mod |ENCODINGS|
encoding[i] = ENCODINGS[idx]
]]></artwork>
        <t>where <tt>H</tt> is the session's negotiated hash family (see
<xref target="hash-family"/> and <xref target="crypto-profile"/>) and <tt>cover_batch_nonce</tt> is a
fresh 16-byte random value generated at the start of each cover
batch. The receiver, holding the session topology, discards cover
packets at the witness validation stage and never invokes the
polyglot decoder.</t>
      </section>
      <section anchor="security-property">
        <name>Security Property</name>
        <t>An adversary observing the complete on-wire stream on any single substrate
sees a deliberately heterogeneous mix of presentation shapes. Without the
session topology AND the cover-batch nonce, the adversary cannot determine:</t>
        <ol spacing="normal" type="1"><li>
            <t>which packets are real vs cover;</t>
          </li>
          <li>
            <t>what presentation encoding the real packets use (they appear as one
base64 stream among seven others);</t>
          </li>
          <li>
            <t>whether the substrate carries a single coherent transmission, multiple
overlapping transmissions, or unrelated traffic.</t>
          </li>
        </ol>
        <t>This extends the indistinguishability guarantee of cover packets from the
cryptographic witness layer (which holds against any adversary) up to the
substrate-encoding fingerprint layer (which holds against the specific
adversary class of traffic-shape analysers).</t>
      </section>
    </section>
    <section anchor="relay-manifest">
      <name>External Relay Manifest</name>
      <t>A Chorale deployment achieves substrate diversity (claim 6 of the
underlying patent) by dispatching atomic packets across multiple
physical network paths. This section specifies an <bcp14>OPTIONAL</bcp14> registry
mechanism by which a deployment may declare and discover the relay
endpoints that constitute its substrate fabric.</t>
      <section anchor="relay-registry-schema">
        <name>Relay Registry Schema</name>
        <t>Each registered relay endpoint is described by a JSON object with the
following fields:</t>
        <artwork><![CDATA[
{
  "relay_code":   "<operator-assigned slug>",
  "provider":     "cloudflare" | "fly" | "render" | "vercel" |
                  "aws" | "gcp" | "azure" | "self-hosted",
  "region":       "<provider-specific region identifier>",
  "jurisdiction": "<ISO 3166-1 alpha-2>",
  "substrates":   [ "fibre" | "satellite" | "microwave" |
                    "mesh" | "sneakernet" | "sovereign" ],
  "forward_url":  "https://<host>/forward",
  "shared_secret_hash": "<hex digest of HMAC key under the
                        session's negotiated hash family>"
}
]]></artwork>
        <t>The <tt>shared_secret_hash</tt> length and computation <bcp14>MUST</bcp14> match the
negotiated hash family of <xref target="hash-family"/>: 64 hex characters for
SHA-256, 128 hex characters for SHA-512 and SHA3-512.</t>
      </section>
      <section anchor="dispatch-hmac-authentication">
        <name>Dispatch HMAC Authentication</name>
        <t>Every dispatch from sender to a registered relay endpoint <bcp14>MUST</bcp14> carry an
HMAC signature in the <tt>X-Chorale-Signature</tt> header, computed under the
HMAC primitive <xref target="FIPS198-1"/> parameterised by the session's negotiated
hash family of <xref target="hash-family"/>, over the request body using the shared
secret recorded against the relay's registry entry. Relays <bcp14>MUST</bcp14> reject
requests whose signature does not match. This prevents a registered
relay from being weaponised as an injection point by third parties.</t>
      </section>
      <section anchor="substrate-diversity-constraint">
        <name>Substrate Diversity Constraint</name>
        <t>A deployment compliant with this extension <bcp14>SHOULD</bcp14> maintain a registry
such that the active relays collectively satisfy:</t>
        <ul spacing="normal">
          <li>
            <t>at least three (3) distinct <tt>jurisdiction</tt> values;</t>
          </li>
          <li>
            <t>at least three (3) distinct <tt>provider</tt> values;</t>
          </li>
          <li>
            <t>coverage of every <tt>substrate</tt> type the deployment dispatches across.</t>
          </li>
        </ul>
        <t>When these conditions hold, the security property of <xref target="flight"/> is
strengthened from "no single substrate carries more than the threshold"
to "no single jurisdiction and no single provider observes more than
the threshold", which is the operational form of the property that
defends against state-level adversaries with single-jurisdiction reach.</t>
      </section>
    </section>
    <section anchor="reproducibility">
      <name>Reproducibility Surface</name>
      <t>This section specifies an <bcp14>OPTIONAL</bcp14> test-vector reproducibility surface
that permits any third party to verify implementation conformance to
this draft without requiring access to implementation source code.</t>
      <section anchor="canonical-test-vectors">
        <name>Canonical Test Vectors</name>
        <t>A conforming implementation <bcp14>SHOULD</bcp14> ship a canonical set of deterministic
test vectors. Each test vector specifies:</t>
        <ul spacing="normal">
          <li>
            <t><tt>id</tt>               -  globally unique identifier (e.g. <tt>CHORALE-VEC-001</tt>)</t>
          </li>
          <li>
            <t><tt>hash_family</tt>      -  the negotiated hash family identifier</t>
          </li>
          <li>
            <t><tt>N</tt>, <tt>K</tt>           -  topology parameters</t>
          </li>
          <li>
            <t><tt>pad_hex</tt>          -  fixed pad bytes as hex</t>
          </li>
          <li>
            <t><tt>topology_edges</tt>   -  fixed neighbour graph</t>
          </li>
          <li>
            <t><tt>plaintext</tt>        -  fixed input string</t>
          </li>
          <li>
            <t><tt>expected.merkle_root</tt>          -  digest of plaintext UTF-8 bytes
                                 under the vector's hash family</t>
          </li>
          <li>
            <t><tt>expected.pad_consumed</tt>         -  pad bytes consumed by the encoding</t>
          </li>
          <li>
            <t><tt>expected.real_body_sha</tt>        -  per-node body digest array under
                                 the vector's hash family</t>
          </li>
          <li>
            <t><tt>expected.real_neighbour_witness</tt>  -  per-node witness array under
                                    the vector's hash family</t>
          </li>
        </ul>
        <t>Three minimum vectors are <bcp14>RECOMMENDED</bcp14> per hash family:</t>
        <ul spacing="normal">
          <li>
            <t>one short vector with <tt>N=4</tt>, <tt>K=2</tt>, plaintext length &lt;= <tt>N</tt>;</t>
          </li>
          <li>
            <t>one medium vector with <tt>N=8</tt>, <tt>K=3</tt>, plaintext length &gt; <tt>N</tt>;</t>
          </li>
          <li>
            <t>one edge-case vector with plaintext length &lt; <tt>N</tt>.</t>
          </li>
        </ul>
      </section>
      <section anchor="cross-implementation-agreement">
        <name>Cross-Implementation Agreement</name>
        <t>A conforming implementation <bcp14>SHOULD</bcp14> expose a public reproducibility
endpoint at which:</t>
        <ol spacing="normal" type="1"><li>
            <t>Every test vector is executed against the server-side implementation;</t>
          </li>
          <li>
            <t>The same vectors are re-executable against a browser-resident
implementation (or other independent implementation);</t>
          </li>
          <li>
            <t>A cross-implementation agreement matrix is rendered showing, for
each vector, whether the implementations produced byte-identical
outputs at every protocol layer (merkle root, body hashes,
neighbour witnesses, reassembled plaintext) under the vector's
declared hash family.</t>
          </li>
        </ol>
        <t>Any party may re-execute the agreement matrix without source-code access.
This establishes claim-to-code traceability for the protocol's named
properties and is the foundation on which third-party security review
is conducted.</t>
      </section>
    </section>
    <section anchor="comparison-with-related-work">
      <name>Comparison with Related Work</name>
      <t><xref target="RFC9846"/> (TLS 1.3) provides confidentiality based on
computational hardness assumptions vulnerable to retrospective
decryption under quantum capability. Chorale attains the same
<tt>computational</tt> regime on its derived-keystream path, and admits an
<tt>everlasting</tt> or <tt>information-theoretic</tt> regime at the payload layer
where the keying conditions of <xref target="regime-floor"/> are met, which TLS
does not define.</t>
      <t><xref target="Chaum1981"/> established the mixnet as the response to traffic
analysis. The Tor protocol, in that lineage, provides pseudonymity
but remains subject to traffic analysis and defines no payload-layer
confidentiality regime above the computational.</t>
      <t><xref target="Loopix2017"/> provides cover traffic at the timing level; Chorale
extends cover indistinguishability to the integrity-witness structure
level.</t>
      <t><xref target="Shamir1979"/> secret sharing and <xref target="ReschPlank2011"/> AONT-RS encode
share ordering or share identity within the share; Chorale moves
share ordering into the per-session secret Topology Graph and
transmits no ordering information on the wire.</t>
      <t><xref target="RSW1996"/> time-lock puzzles and <xref target="Wesolowski2018"/> VDFs are used
per-message in prior art; Chorale applies them per-packet to
multiply adversary cost by N while preserving receiver
parallelisability.</t>
    </section>
    <section anchor="implementation-considerations">
      <name>Implementation Considerations</name>
      <t>Implementations should consider the following operational matters:</t>
      <t><strong>Endpoint hardware tamper evidence.</strong> Endpoints should be commissioned
with optical microscopy verification of the silicon dies of the
multi-sovereign entropy generator and the Verifiable Delay Function
evaluator against vendor reference images. Electromagnetic emanation
profiling against a baseline recorded in a certified anechoic chamber
should be performed prior to each session.</t>
      <t><strong>Pad replenishment.</strong> The one-time pad is exhausted by transmission.
Implementations should provide for pad replenishment via further
co-presence events or via Quantum Key Distribution at the rate
required for the projected message volume.</t>
      <t><strong>Receiver hardware parallelism.</strong> The Verifiable Delay Function
evaluator at the receiver should be capable of parallel operation
across multiple physical processors to minimise reassembly latency.</t>
      <t><strong>Substrate diversity.</strong> Implementations should ensure that the
substrates available to a session include at least one substrate
that is resilient to anticipated adversary disruption capabilities
(e.g., undersea cable cut, satellite spot-beam interception,
microwave path obstruction).</t>
    </section>
    <section anchor="covert-posture">
      <name>Covert Posture and Constant-Rate Pacing</name>
      <t>A session <bcp14>MAY</bcp14> set a <tt>covert</tt> flag. When set, the stealth defence set  - 
constant-rate pacing, cover-continuity, decoy pairing, and rotating
rendezvous  -  is enabled at ANY posture, decoupling undetectability from the
threat tier. A non-covert session behaves exactly as in prior revisions.</t>
      <t>Under constant-rate pacing the endpoint emits ONE fixed-size cell on a fixed
cadence, whether or not real data is queued. A real cell carries an
encrypt-then-MAC frame body; a chaff cell is uniform random of the identical
wire size. Real and chaff cells are byte-for-byte indistinguishable to any
party without the session key: distinguishing a real cell from chaff reduces
to forging the cell MAC. Timing analysis, volume analysis, and the
"who is active when" signal are therefore removed from the intra-stream layer.
Because the defence relies on the stream being message-independent, the stream
<bcp14>MUST</bcp14> run continuously; starting or stopping it in step with message activity
would reintroduce the signal it removes.</t>
    </section>
    <section anchor="regime-floor">
      <name>Confidentiality-Regime Floor and Corroboration</name>
      <t>A session achieves one of three confidentiality regimes, determined by its
keying method: <tt>computational</tt> (client hybrid-PQC handoff or server-generated
pad), <tt>everlasting</tt> (bounded-storage class), or <tt>information-theoretic</tt>
(an unconditional co-presence or QKD strand). A policy <bcp14>MAY</bcp14> declare a
<tt>requiredFloor</tt>. Every send path enforces the floor identically and FAILS
CLOSED: a send is refused when the regime relied upon is below the floor, when
the policy is malformed, or on infrastructure error.</t>
      <t>A raised regime (<tt>everlasting</tt> or <tt>information-theoretic</tt>) is RELIED upon only
when the co-presence or bounded-storage strand is INDEPENDENTLY CORROBORATED  - 
for example by completing the out-of-band SAS confirmation of <xref target="hash-family-sas"/>.
An un-corroborated claim does not inflate the regime; it down-rates to
<tt>computational</tt>. A pad held by the server (persisted server-side) <bcp14>MUST NOT</bcp14>
claim a regime above <tt>computational</tt> under any declared co-presence class,
because such a pad cannot deliver unconditional secrecy against the server or a
subsequent compulsion of it. Only a client-keyed session, whose pad never
leaves the endpoint, may claim a raised regime.</t>
    </section>
    <section anchor="cover-ratio">
      <name>Negotiated Cover Ratio</name>
      <t>The ratio of cover packets to real packets is operator-negotiable, carried in
the session hardening policy, with a default of three cover packets per real
packet. A negotiated ratio applies for the session lifetime.</t>
    </section>
    <section anchor="beacon-quorum">
      <name>Beacon Quorum and Trust-Root Pluralisation</name>
      <t>Where a session binds to public randomness (for rendezvous or epoch binding),
it <bcp14>MAY</bcp14> draw that randomness from a quorum of independent public beacons rather
than a single beacon. The epoch value is derived by HKDF over all contributions
under a robust combiner with three properties:</t>
      <ol spacing="normal" type="1"><li>
          <t>UNPREDICTABLE-IF-ANY-ONE-HONEST: an adversary controlling all but one
contributor can neither bias nor predict the output while one honest
contributor remains.</t>
        </li>
        <li>
          <t>QUORUM-GATED: a minimum number of distinct beacons <bcp14>MUST</bcp14> be present, so a
withheld beacon below quorum is detected rather than silently trusted.</t>
        </li>
        <li>
          <t>VERIFIER-TRANSPARENT: each combine emits a transcript hash binding the exact
contributions, so any relying party recomputes and checks the epoch value
without trusting the operator.</t>
        </li>
      </ol>
      <t>An <bcp14>OPTIONAL</bcp14> private commit-reveal mix folds endpoint entropy into the epoch
value so that it is additionally unpredictable to a public observer, with a
last-actor grinding attempt detected as a commit-reveal mismatch. Combined with
the sequential-hash time-lock of <xref target="vdf"/>  -  which provides delay without trusting
any clock  -  the trust-root surface carries no single load-bearing root: not the
beacon (plural under a combiner), not the clock (delay function), and not the
verifier (public transcript).</t>
    </section>
    <section anchor="hallmark-binding">
      <name>Positive-Attestation Binding</name>
      <t>This section is informative. The quorum-beacon value of <xref target="beacon-quorum"/> <bcp14>MAY</bcp14>
additionally serve as the moving-target binding for a co-deployed positive-
attestation engine, so that a single trust-root-plural randomness spine serves
both this transport and the attestation layer. The attestation layer is out of
scope for the normative behaviour of this protocol.</t>
    </section>
    <section anchor="assurance">
      <name>Operator Authentication and Assurance Binding</name>
      <t>Chorale binds each message or session to a live operator authenticator (the
hardware-presence binding of the co-presence and endpoint model). This section
specifies how a deployment REQUIRES and BINDS the ASSURANCE LEVEL of that
authentication  -  authenticator assurance, biometric liveness, and identity
proofing  -  WITHOUT the protocol performing biometric matching. All biometric
processing remains within the operator's certified authenticator or
presentation-attack-detection subsystem; the protocol binds only the resulting
signed ATTESTATIONS, never a biometric template. This aligns a deployment with
the authenticator- and identity-assurance levels of the applicable scheme (for
United States federal use, NIST SP 800-63) and, where biometrics are used, with
the presentation-attack-detection requirements of ISO/IEC 30107.</t>
      <section anchor="assurance-factors">
        <name>Assurance Factors</name>
        <t>An operator <bcp14>MAY</bcp14> present one or more factor attestations with a send. Each
attestation is an opaque, verifier-checkable evidence object bound to a fresh
challenge:</t>
        <ul spacing="normal">
          <li>
            <t>possession: the authenticator produced an assertion (a token was held).</t>
          </li>
          <li>
            <t>user_verification: the authenticator performed a biometric or PIN check (the
WebAuthn user-verification result), mapping to an authenticator-assurance level.</t>
          </li>
          <li>
            <t>liveness: a certified subsystem attests presentation-attack detection at a
stated level.</t>
          </li>
          <li>
            <t>identity_proofing: a reference to an identity-proofing result at a stated level.</t>
          </li>
        </ul>
        <t>An attestation <bcp14>MUST</bcp14> carry the fresh challenge it answers. The relying endpoint
<bcp14>MUST</bcp14> issue a single-use challenge and <bcp14>MUST</bcp14> reject any attestation whose challenge
does not match the issued value, so a recorded assertion cannot be replayed
within or across sessions.</t>
      </section>
      <section anchor="assurance-floor">
        <name>Assurance Floor</name>
        <t>A session policy <bcp14>MAY</bcp14> declare an assurance floor: a minimum number of DISTINCT
independent factors (a quorum), <bcp14>OPTIONAL</bcp14> pinned factors, and <bcp14>OPTIONAL</bcp14> minimum
authenticator-, liveness-, and identity-assurance levels. The send path <bcp14>MUST</bcp14>
refuse to transmit when the floor is not met, in the same fail-closed manner as
the confidentiality-regime floor (<xref target="regime-floor"/>), and <bcp14>MUST</bcp14> refuse when the
policy is malformed or when a policy is present but no attestation is supplied.</t>
        <t>Consistent with the protocol's principle that no single element is individually
load-bearing, the floor is by default a THRESHOLD over independent factors: a
policy that requires a quorum without pinning any specific factor is met by any
sufficient subset, so the loss of a single factor does not by itself deny an
otherwise-authorised session. A deployment <bcp14>MAY</bcp14> pin a specific factor where its
regime demands it; that is an explicit operator choice and does not change the
threshold semantics for the remaining factors.</t>
      </section>
      <section anchor="binding-and-privacy">
        <name>Binding and Privacy</name>
        <t>Admitted factor attestations are folded into the local key derivation in a
DETERMINISTIC order, so the binding is independent of the order in which factors
are presented and remains sound if any one folded attestation is unpredictable to
the adversary  -  the same robust-combiner property as the beacon quorum of
<xref target="beacon-quorum"/>. Only a hash COMMITMENT over the folded attestations is written
to any sealed record; the attestation bytes and any personally identifying
material are never persisted. A session with no assurance policy is unaffected by
this section.</t>
      </section>
    </section>
    <section anchor="architecture">
      <name>Deployment Architecture and Integration Context</name>
      <t>This section is informative. It describes the environment in which a Chorale
endpoint is deployed and the interfaces between the Chorale transport and the
surrounding components of the reference platform. None of it is required for two
conforming Chorale endpoints to interoperate; it is provided so that an
implementer and a security reviewer understand the trust boundaries the protocol
assumes and the mitigations available for the residual risks in the Security
Considerations.</t>
      <section anchor="hw-model">
        <name>Endpoint and Hardware Model</name>
        <t>Chorale is designed to run on commodity hardware. It requires no hardware
security module, trusted platform module, or trusted execution environment to
operate; the reference implementation runs in an ordinary browser and in a
server runtime. Specialised hardware, where present, strengthens key custody but
is never a precondition for the protocol's security properties.</t>
        <t>Three endpoint mechanisms bound the risk that the endpoint  -  the honest primary
attack surface once the wire is defeated  -  presents.</t>
        <ul spacing="normal">
          <li>
            <t>Randomness. Key, pad, share, and nonce generation draws from a multi-source
generator whose output is the exclusive-or of the system generator and an
independently seeded HMAC-DRBG, admitted only after a NIST SP 800-90B entropy
health test. The exclusive-or of independent sources is sound if any one source
is sound, so a single biased or backdoored generator can neither control nor
predict the output.</t>
          </li>
          <li>
            <t>Pad at rest. The one-time-pad reservoir is never at rest in exportable
plaintext. Pad bytes are wrapped under a non-exportable, hardware-backed key,
and pad regions are handed to a consumer under a draw-and-destroy discipline
that zeroes the source immediately after use, minimising residency and removing
the cloud-profile-sync exposure a non-live adversary would otherwise harvest.</t>
          </li>
          <li>
            <t>Key custody. The pad <bcp14>MAY</bcp14> additionally be bound to a hardware token the operator
already carries (for example a WebAuthn PRF or challenge-response token) and
split under a threshold secret-sharing scheme across independent stores:
endpoint storage, a hardware-held share, and a server-blinded share. Any
sub-threshold subset is information-theoretically empty; a honey-wrap ensures a
sub-threshold or wrong-token unwrap yields a plausible decoy rather than an
error, denying an offline oracle.</t>
          </li>
        </ul>
        <t>These mechanisms protect key material at rest and past traffic after a later
endpoint image; they do not defend a live endpoint compromised while composing a
message. That residual is bounded, not removed, by the session key ratchet.</t>
      </section>
      <section anchor="sovereignty">
        <name>Data Sovereignty and Jurisdictional Diversity</name>
        <t>Chorale is sovereignty-preserving in three concrete senses. First, in the
client-keyed regime (<xref target="regime-floor"/>) the pad never leaves the endpoint and the
relay and server never observe plaintext, so no operator, host, or jurisdiction
holds the material required to read traffic. Second, packet flight is distributed
across substrates in distinct jurisdictions (<xref target="relay-manifest"/>), so no single
jurisdiction observes the whole of a transmission and compelling any one
substrate operator yields only cover-indistinguishable fragments. Third, the
assignment of substrate and jurisdiction to a session is operator-controlled
policy, so a deployment <bcp14>MAY</bcp14> enforce data-residency and routing constraints
without changing the protocol. The server-blind relay commitment layer
(<xref target="commitment"/>) lets each relay attest that it forwarded what it received
without being able to read it.</t>
      </section>
      <section anchor="membrane">
        <name>Cryptographic-Sovereignty Boundary and Breach Containment</name>
        <t>A Chorale endpoint <bcp14>MAY</bcp14> be deployed behind a cryptographic-sovereignty boundary
that binds keys to the boundary and permits a relying party to verify sealed
records with only the artefact and a published key, without trusting the platform
that produced them; the platform holds only a public fingerprint of a secret,
never the secret itself. The boundary carries a breach-containment sentinel: on a
seal-verification failure, a detected retrospective drift, or an access denial,
the sentinel commits a structured breach event  -  anchored to an external ledger  - 
quarantines the affected session, and locks the affected principal. Each step of
the containment path is best-effort and independent, so a breach response
completes even if an individual subsystem is unavailable, and an incomplete
on-chain anchor is retried on the next surveillance cycle. From the protocol's
perspective this is a fail-closed response to boundary compromise: a boundary
that can no longer prove its own integrity is rendered inert rather than trusted.</t>
      </section>
      <section anchor="vault">
        <name>Threshold Key Custody and Time-Lock</name>
        <t>The verifiable-delay sealing of <xref target="vdf"/> is one half of a custody model in which
single-use key material is additionally held under threshold secret-sharing and
released only after an unavoidable delay. Material under time-lock cannot be
reached or re-encrypted before its release, which removes the leverage a
ransomware adversary would otherwise obtain over an endpoint's key store, and the
threshold split ensures no single store holds a usable key.</t>
      </section>
      <section anchor="display">
        <name>Display and Anti-Capture of Delivered Plaintext</name>
        <t>After Chorale delivers plaintext to an endpoint, the last exposure is the display
itself  -  a screen capture or a camera photograph of the rendered content. A
Chorale deployment <bcp14>MAY</bcp14> render delivered content through an optical-capture-
resistant, ephemeral display that reconstructs the content only transiently
across temporal shares, so a single captured frame does not yield it, and <bcp14>MAY</bcp14>
seed the surface with watermarked decoy assets whose leak is self-incriminating.
This closes the delivery path against an observer who has defeated neither the
cryptography nor the endpoint but simply photographs the screen.</t>
      </section>
      <section anchor="composition">
        <name>Composition and Assembly</name>
        <t>Chorale is one component of a composed platform assembled over a shared
sealed-attestation substrate. A composition layer maintains a registry of the
platform's functional components, a unified verdict envelope, and a cross-
component routing map, so that Chorale transport can be invoked by any platform
surface requiring interception-resistant delivery without that surface
re-implementing the protocol. This is deployment context and imposes no
requirement on the wire protocol.</t>
      </section>
      <section anchor="reality">
        <name>Reality Attestation</name>
        <t>The beacon-quorum value of <xref target="beacon-quorum"/> is shared with a reality-attestation
layer that binds an artefact's claimed time and place to a quorum of independent
public physical registers  -  for example mains-frequency continuity, provenance
markers, and public randomness beacons  -  under a not-before randomness
commitment, sealing the result as an independently replayable record. Divergence
among the quorum of sealed sensors is treated as an adversary-adaptation signal.
This layer is the positive-attestation counterpart to Chorale's confidentiality:
Chorale establishes that a message could not be read; the reality-attestation
layer establishes that an artefact's asserted origin could not have been
fabricated below a stated cost. The two share one trust-root-plural randomness
spine and are otherwise independent.</t>
      </section>
    </section>
    <section anchor="security-considerations">
      <name>Security Considerations</name>
      <t>The Chorale Protocol is designed to defend against the following
adversary capabilities:</t>
      <ol spacing="normal" type="1"><li>
          <t><strong>Wire-level interception</strong> of all packets across all substrates
simultaneously. Resisted at the strength of the regime in force
under <xref target="regime-floor"/>. Under the <tt>information-theoretic</tt> regime
the adversary obtains ciphertext carrying no information about the
plaintext. Under <tt>everlasting</tt> or <tt>computational</tt> the payload rests
on the stated assumption and the guarantee is bounded by it.</t>
        </li>
        <li>
          <t><strong>Wire-level packet injection or modification</strong>. Defeated by
cascade Neighbour-Witness integrity over the hidden Topology
Graph; any modification cascades to invalidate K+1 witnesses.</t>
        </li>
        <li>
          <t><strong>Traffic analysis</strong> including timing correlation and packet-count
correlation. Defeated by Cover Packets indistinguishable from
Atomic Packets at the integrity-witness level, combined with
prime-modular timing dispatch.</t>
        </li>
        <li>
          <t><strong>Retrospective decryption</strong> by adversaries with future quantum
capability. Defeated under the <tt>information-theoretic</tt> regime,
which is unconditionally secure against an unbounded adversary.
Under <tt>everlasting</tt> the payload is secure against an adversary
bounded in storage at the declared bound. Under <tt>computational</tt>
harvest-now-decrypt-later is mitigated by the post-quantum
handshake and is not eliminated.</t>
        </li>
        <li>
          <t><strong>Single-substrate compromise</strong> including cable taps, satellite
spot-beam interception, and substrate denial-of-service.
Defeated by multi-substrate flight requiring K-of-M shares across
diverse path types.</t>
        </li>
        <li>
          <t><strong>Single-entropy-source compromise</strong> including supply-chain
compromise of an entropy generator or sovereign-government
coercion of an entropy source operator. Defeated by triple-
sovereign-XOR pad composition.</t>
        </li>
        <li>
          <t><strong>Pad reuse</strong> including operator error and deliberate reuse.
Defeated by the Tombstone Ledger, which structurally rejects
reuse at the receiver.</t>
        </li>
        <li>
          <t><strong>Endpoint coercion</strong> of a sender operator. Mitigated by the
duress codeword mechanism, which silently flags receiver-side
audit to a duress condition.</t>
        </li>
      </ol>
      <t>The protocol does NOT defend against the following residual attack
surfaces, which are addressed by operational and hardware measures
outside the scope of this specification:</t>
      <ul spacing="normal">
        <li>
          <t>Endpoint compromise of either the sender or the receiver
(mitigated by hardware tamper-evidence, TEMPEST-certified
operating environments, and Shamir-style split custody of the
Session Anchor);</t>
        </li>
        <li>
          <t>Side-channel attacks on the Verifiable Delay Function evaluator
hardware (mitigated by constant-time implementations and
hardware-pinned VDF chips);</t>
        </li>
        <li>
          <t>Rubber-hose attacks producing the Session Anchor (mitigated by
split custody and non-extradition-jurisdiction custodian
distribution).</t>
        </li>
      </ul>
      <t>Implementers <bcp14>MUST</bcp14> consider these residual attack surfaces and
implement appropriate operational and hardware measures.</t>
      <section anchor="sec-mitm-downgrade">
        <name>Hash Family Negotiation Under Active MITM</name>
        <t>An active man-in-the-middle who intercepts the rendezvous payload
can attempt a hash-family downgrade by rewriting one peer's
advertised list before forwarding it to the other (and symmetrically
on the return). For example, Alice advertises
<tt>[sha3-512, sha-512]</tt> and Bob advertises
<tt>[sha3-512, sha-512, sha-256]</tt>. An MITM strips <tt>sha3-512</tt> and
<tt>sha-512</tt> from Alice's view of Bob's advertisement (or vice versa),
forcing the intersection-and-strongest algorithm to land on
<tt>sha-256</tt> so the MITM can later mount cryptanalytic attacks at the
weaker security margin.</t>
        <t>Chorale defends against this attack by binding the negotiated
<tt>hash_family</tt> byte into the Short Authentication String preimage
recited out-of-band at handshake completion, per
<xref target="hash-family-sas"/>. The SAS digest is computed as:</t>
        <artwork><![CDATA[
SAS_digest = H_negotiated( hash_family || "|" || SAS_words )
]]></artwork>
        <t>An MITM who forces Alice to negotiate <tt>sha-256</tt> while leaving Bob
on the belief that they have negotiated <tt>sha-512</tt> produces SAS
digests under different families with different family-byte
preimages. When Alice reads her SAS aloud and Bob compares it
against his own, the words diverge and the divergence is detected
before any session traffic is transmitted. The session is then
aborted at the operator layer.</t>
        <t>This defence is structural: it does not require the operators to
understand cryptography. They follow the existing SAS recital
discipline (read aloud, compare, abort on mismatch); the
<tt>hash_family</tt> byte in the preimage is what makes that discipline
suffice against the downgrade attack.</t>
        <t>The defence depends on the out-of-band channel used for SAS recital
being secret from the MITM. A telephone call, a video-call audio
channel, or a physically-co-located reading all suffice. Any
channel the MITM can also rewrite (e.g., an in-band text message
without operator visual identity verification) does NOT suffice
and <bcp14>MUST NOT</bcp14> be used for SAS recital.</t>
      </section>
      <section anchor="hash-family-selection">
        <name>Hash Family Selection</name>
        <t>This subsection records the rationale for the hash-family choices
of <xref target="hash-family"/>.</t>
        <t><tt>sha-512</tt> is the <bcp14>RECOMMENDED</bcp14> default for new deployments. Relative
to the v-01 baseline of <tt>sha-256</tt>, the wider internal state of
SHA-512 provides a larger collision-resistance band against any
future cryptanalytic advance in the SHA-2 family's compression
function and against quantum-adversary Grover-search applied to
preimage finding. The 256-bit effective post-quantum-Grover
preimage strength of SHA-512 matches the security band of an
unexpanded pad and of post-quantum signature schemes
parameterised at the 256-bit security level.</t>
        <t><tt>sha3-512</tt> <bcp14>SHOULD</bcp14> be preferred where deployment posture demands
diversity at the cryptographic-construction level. SHA-512 and
SHA-256 share the Merkle-Damgard construction lineage; any
cryptanalytic discovery affecting the SHA-2 compression function
collapses both. SHA3-512, built on the Keccak sponge construction,
is structurally independent and would not be co-affected by such
a discovery. The same logic that motivates the multi-source
entropy generator at the pad layer motivates SHA3-512 as a hedge
at the hash-family layer.</t>
        <t><tt>sha-256</tt> is RETAINED only as a backward-compatible legacy mode
for resource-constrained devices that cannot accommodate the
larger block size of SHA-512. New deployments <bcp14>SHOULD NOT</bcp14> advertise
<tt>sha-256</tt> only. Mixed-corridor deployments where one endpoint is
constrained <bcp14>MAY</bcp14> accept <tt>sha-256</tt> negotiation for that corridor.</t>
      </section>
      <section anchor="allied-artefact-crypto-profile-binding-rides-opaque-through-chorale">
        <name>Allied Artefact Crypto-Profile Binding Rides Opaque Through Chorale</name>
        <t>The Chorale Protocol transmits Allied verification-artefact crypto-
profile algorithm tags as opaque bytes within the encrypted payload.
The protocol does not interpret, validate, or constrain the per-
artefact hash family carried inside the payload. An Allied corridor
operating under the Chorale Protocol may carry artefacts tagged
under any algorithm in the Allied crypto-profile registry, and the
Chorale transport layer makes no inference on those tags.</t>
        <t>This separation is deliberate. The Chorale hash-family negotiation
of <xref target="hash-family"/> governs the transport-layer primitives that
defend the wire: HKDF derivation, VDF sealing, commitment-layer
digests, and per-packet HMAC tags. The Allied artefact crypto-
profile, defined and managed in the Allied verification platform
specification, governs the per-artefact content-hash algorithm used
for ingestion, evidence-chain commitment, and envelope sealing.
These two surfaces are intentionally orthogonal. A constrained-
device corridor negotiating <tt>sha-256</tt> for Chorale <bcp14>MAY</bcp14> carry an
Allied artefact whose internal crypto-profile is <tt>sha3-512</tt>. A
sovereign-corridor session negotiating <tt>sha3-512</tt> for Chorale <bcp14>MAY</bcp14>
carry artefacts tagged <tt>sha-256</tt> from a partner-tenant origin.</t>
        <t>Implementations <bcp14>MUST NOT</bcp14> cross-validate the Chorale negotiated
hash family against the Allied artefact crypto-profile. The two
surfaces are governed by distinct registries and have distinct
audit trails.</t>
      </section>
      <section anchor="constant-time-comparison-and-non-persistence">
        <name>Constant-Time Comparison and Non-Persistence</name>
        <t>All secret-dependent comparisons  -  neighbour-witness verification, dispatch
HMAC verification, commitment and Merkle comparison, and duress-codeword
comparison  -  are performed in constant time, so no secret is leaked through
comparison timing. The relay and receiver persist neither plaintext, nor a
plaintext-derived commitment, nor any distinguishable duress marker. The stored
record of a send made under a duress codeword is indistinguishable from that of
a normal send; the duress signal is surfaced only out-of-band to the recipient
endpoint. This preserves the deniability on which coercion resistance depends.</t>
      </section>
      <section anchor="empirical-indistinguishability-of-the-constant-rate-stream">
        <name>Empirical Indistinguishability of the Constant-Rate Stream</name>
        <t>The real-versus-decoy indistinguishability of the constant-rate stream
(<xref target="covert-posture"/>) is measured rather than asserted. Over a large sample of
real and chaff cells drawn from the reference implementation, a panel of
statistical distinguishers and a trained classifier separate real from chaff
cells with an area-under-curve within statistical noise of the 0.5 coin-flip
baseline, while the same distinguishers separate a deliberately unprotected
control format at 1.0, establishing that the null result reflects genuine
indistinguishability and not an insensitive test. Pooled keystream entropy is
within measurement error of the 8.0 bits-per-byte uniform ideal. The method and
figures are recorded in the implementation's assurance harnesses.</t>
      </section>
      <section anchor="endpoint-compromise-and-hardware-assumptions">
        <name>Endpoint Compromise and Hardware Assumptions</name>
        <t>Chorale assumes no specialised hardware for its security properties; see
<xref target="hw-model"/>. The endpoint is the primary residual attack surface once the wire
is defeated. The at-rest pad protection, multi-source randomness, and threshold
hardware-bound custody of <xref target="hw-model"/> reduce the value of an endpoint image and
deny an offline oracle, but a live endpoint compromised during composition is a
stated ceiling of the protocol, bounded by the session key ratchet and, at the
display, by the anti-capture rendering of <xref target="display"/>. Breach of a cryptographic-
sovereignty boundary, where deployed, triggers the fail-closed containment of
<xref target="membrane"/>.</t>
      </section>
    </section>
    <section anchor="iana-considerations">
      <name>IANA Considerations</name>
      <t>This document requests the establishment of a "Chorale Hash Family"
registry, with the following initial registrations:</t>
      <table>
        <thead>
          <tr>
            <th align="left">Identifier</th>
            <th align="left">Underlying primitive</th>
            <th align="left">Status</th>
            <th align="left">Reference</th>
          </tr>
        </thead>
        <tbody>
          <tr>
            <td align="left">sha-256</td>
            <td align="left">SHA-256 <xref target="FIPS180-4"/></td>
            <td align="left">Legacy</td>
            <td align="left">This memo</td>
          </tr>
          <tr>
            <td align="left">sha-512</td>
            <td align="left">SHA-512 <xref target="FIPS180-4"/></td>
            <td align="left">Recommended (default)</td>
            <td align="left">This memo</td>
          </tr>
          <tr>
            <td align="left">sha3-512</td>
            <td align="left">SHA3-512 <xref target="FIPS202"/></td>
            <td align="left">Optional</td>
            <td align="left">This memo</td>
          </tr>
        </tbody>
      </table>
      <t>The registration policy for new entries is "Specification Required"
per BCP 26. New entries <bcp14>MUST</bcp14> identify the underlying hash primitive,
the digest length, and the status (Legacy, Recommended, Optional).
New entries <bcp14>MUST</bcp14> be assessed against the digest-length and effective
preimage-strength requirements stated in <xref target="hash-family"/> and
<xref target="commitment"/>.</t>
      <t>A future version of this document may additionally register:</t>
      <ul spacing="normal">
        <li>
          <t>A media type "application/chorale-packet" for Atomic Packets and
Cover Packets;</t>
        </li>
        <li>
          <t>A URI scheme "chorale:" for Session Anchor references;</t>
        </li>
        <li>
          <t>A registry of named cryptographic profiles for the algorithm
selection described in <xref target="crypto-profile"/>.</t>
        </li>
      </ul>
    </section>
    <section anchor="related-standards">
      <name>Related Standards</name>
      <t>The Chorale Protocol is part of a family of standards drafted by
the same author for verification-anchored infrastructure:</t>
      <ul spacing="normal">
        <li>
          <t><xref target="SCITT-ARP"/> defines an attestation resolution profile for the
SCITT framework.</t>
        </li>
        <li>
          <t><xref target="E8VS"/> defines a verified profile for the Australian Essential
Eight cyber baseline.</t>
        </li>
        <li>
          <t><xref target="AIGVS"/> defines an AI Governance Verification Scaffold.</t>
        </li>
      </ul>
      <t>The Chorale Protocol references <xref target="RFC3161"/> for trusted-timestamp
anchoring of envelopes, <xref target="RFC5869"/> for HKDF, <xref target="FIPS180-4"/>,
<xref target="FIPS198-1"/>, <xref target="FIPS202"/>, <xref target="FIPS186-5"/>, <xref target="FIPS203"/>, <xref target="FIPS204"/>
for cryptographic primitive selection, and <xref target="RFC9794"/> for
post-quantum hybrid terminology where applicable.</t>
    </section>
  </middle>
  <back>
    <references anchor="sec-combined-references">
      <name>References</name>
      <references anchor="sec-normative-references">
        <name>Normative References</name>
        <reference anchor="RFC2119">
          <front>
            <title>Key words for use in RFCs to Indicate Requirement Levels</title>
            <author fullname="S. Bradner" initials="S." surname="Bradner"/>
            <date month="March" year="1997"/>
            <abstract>
              <t>In many standards track documents several words are used to signify the requirements in the specification. These words are often capitalized. This document defines these words as they should be interpreted in IETF documents. This document specifies an Internet Best Current Practices for the Internet Community, and requests discussion and suggestions for improvements.</t>
            </abstract>
          </front>
          <seriesInfo name="BCP" value="14"/>
          <seriesInfo name="RFC" value="2119"/>
          <seriesInfo name="DOI" value="10.17487/RFC2119"/>
        </reference>
        <reference anchor="RFC8174">
          <front>
            <title>Ambiguity of Uppercase vs Lowercase in RFC 2119 Key Words</title>
            <author fullname="B. Leiba" initials="B." surname="Leiba"/>
            <date month="May" year="2017"/>
            <abstract>
              <t>RFC 2119 specifies common key words that may be used in protocol specifications. This document aims to reduce the ambiguity by clarifying that only UPPERCASE usage of the key words have the defined special meanings.</t>
            </abstract>
          </front>
          <seriesInfo name="BCP" value="14"/>
          <seriesInfo name="RFC" value="8174"/>
          <seriesInfo name="DOI" value="10.17487/RFC8174"/>
        </reference>
        <reference anchor="RFC3161">
          <front>
            <title>Internet X.509 Public Key Infrastructure Time-Stamp Protocol (TSP)</title>
            <author fullname="C. Adams" initials="C." surname="Adams"/>
            <author fullname="P. Cain" initials="P." surname="Cain"/>
            <author fullname="D. Pinkas" initials="D." surname="Pinkas"/>
            <author fullname="R. Zuccherato" initials="R." surname="Zuccherato"/>
            <date month="August" year="2001"/>
            <abstract>
              <t>This document describes the format of a request sent to a Time Stamping Authority (TSA) and of the response that is returned. It also establishes several security-relevant requirements for TSA operation, with regards to processing requests to generate responses. [STANDARDS-TRACK]</t>
            </abstract>
          </front>
          <seriesInfo name="RFC" value="3161"/>
          <seriesInfo name="DOI" value="10.17487/RFC3161"/>
        </reference>
        <reference anchor="RFC9846">
          <front>
            <title>The Transport Layer Security (TLS) Protocol Version 1.3</title>
            <author fullname="E. Rescorla" initials="E." surname="Rescorla"/>
            <date month="July" year="2026"/>
            <abstract>
              <t>This document specifies version 1.3 of the Transport Layer Security (TLS) protocol. TLS allows client/server applications to communicate over the Internet in a way that is designed to prevent eavesdropping, tampering, and message forgery.</t>
              <t>This document obsoletes RFC 8446, which specified TLS 1.3. This document obsoletes RFC 5246 (specifying TLS 1.2) and RFCs 5077, 6961, 7627, and 8422, all of which pertain to TLS 1.2 or earlier, and updates RFCs 5705 and 6066. This document also specifies new requirements for TLS 1.2 implementations.</t>
            </abstract>
          </front>
          <seriesInfo name="RFC" value="9846"/>
          <seriesInfo name="DOI" value="10.17487/RFC9846"/>
        </reference>
        <reference anchor="RFC5869">
          <front>
            <title>HMAC-based Extract-and-Expand Key Derivation Function (HKDF)</title>
            <author fullname="H. Krawczyk" initials="H." surname="Krawczyk"/>
            <author fullname="P. Eronen" initials="P." surname="Eronen"/>
            <date month="May" year="2010"/>
            <abstract>
              <t>This document specifies a simple Hashed Message Authentication Code (HMAC)-based key derivation function (HKDF), which can be used as a building block in various protocols and applications. The key derivation function (KDF) is intended to support a wide range of applications and requirements, and is conservative in its use of cryptographic hash functions. This document is not an Internet Standards Track specification; it is published for informational purposes.</t>
            </abstract>
          </front>
          <seriesInfo name="RFC" value="5869"/>
          <seriesInfo name="DOI" value="10.17487/RFC5869"/>
        </reference>
        <reference anchor="RFC9794">
          <front>
            <title>Terminology for Post-Quantum Traditional Hybrid Schemes</title>
            <author fullname="F. Driscoll" initials="F." surname="Driscoll"/>
            <author fullname="M. Parsons" initials="M." surname="Parsons"/>
            <author fullname="B. Hale" initials="B." surname="Hale"/>
            <date month="June" year="2025"/>
            <abstract>
              <t>One aspect of the transition to post-quantum algorithms in cryptographic protocols is the development of hybrid schemes that incorporate both post-quantum and traditional asymmetric algorithms. This document defines terminology for such schemes. It is intended to be used as a reference and, hopefully, to ensure consistency and clarity across different protocols, standards, and organisations.</t>
            </abstract>
          </front>
          <seriesInfo name="RFC" value="9794"/>
          <seriesInfo name="DOI" value="10.17487/RFC9794"/>
        </reference>
        <reference anchor="FIPS180-4">
          <front>
            <title>Secure Hash Standard (SHS), Federal Information Processing Standards Publication 180-4</title>
            <author>
              <organization>National Institute of Standards and Technology</organization>
            </author>
            <date year="2015" month="August"/>
          </front>
          <seriesInfo name="FIPS" value="180-4"/>
        </reference>
        <reference anchor="FIPS198-1">
          <front>
            <title>The Keyed-Hash Message Authentication Code (HMAC), Federal Information Processing Standards Publication 198-1</title>
            <author>
              <organization>National Institute of Standards and Technology</organization>
            </author>
            <date year="2008" month="July"/>
          </front>
          <seriesInfo name="FIPS" value="198-1"/>
        </reference>
        <reference anchor="FIPS202">
          <front>
            <title>SHA-3 Standard: Permutation-Based Hash and Extendable-Output Functions, Federal Information Processing Standards Publication 202</title>
            <author>
              <organization>National Institute of Standards and Technology</organization>
            </author>
            <date year="2015" month="August"/>
          </front>
          <seriesInfo name="FIPS" value="202"/>
        </reference>
        <reference anchor="FIPS186-5">
          <front>
            <title>Digital Signature Standard (DSS), Federal Information Processing Standards Publication 186-5</title>
            <author>
              <organization>National Institute of Standards and Technology</organization>
            </author>
            <date year="2023" month="February"/>
          </front>
          <seriesInfo name="FIPS" value="186-5"/>
        </reference>
        <reference anchor="FIPS203">
          <front>
            <title>Module-Lattice-Based Key-Encapsulation Mechanism Standard (ML-KEM)</title>
            <author>
              <organization>National Institute of Standards and Technology</organization>
            </author>
            <date year="2024" month="August"/>
          </front>
          <seriesInfo name="FIPS" value="203"/>
        </reference>
        <reference anchor="FIPS204">
          <front>
            <title>Module-Lattice-Based Digital Signature Standard (ML-DSA)</title>
            <author>
              <organization>National Institute of Standards and Technology</organization>
            </author>
            <date year="2024" month="August"/>
          </front>
          <seriesInfo name="FIPS" value="204"/>
        </reference>
      </references>
      <references anchor="sec-informative-references">
        <name>Informative References</name>
        <reference anchor="Shannon1949">
          <front>
            <title>Communication Theory of Secrecy Systems</title>
            <author initials="C. E." surname="Shannon" fullname="Claude Elwood Shannon">
              <organization/>
            </author>
            <date year="1949"/>
          </front>
          <seriesInfo name="Bell System Technical Journal" value="28(4) 656-715"/>
        </reference>
        <reference anchor="Shamir1979">
          <front>
            <title>How to Share a Secret</title>
            <author initials="A." surname="Shamir">
              <organization/>
            </author>
            <date year="1979"/>
          </front>
          <seriesInfo name="Communications of the ACM" value="22(11) 612-613"/>
        </reference>
        <reference anchor="Chaum1981">
          <front>
            <title>Untraceable Electronic Mail, Return Addresses, and Digital Pseudonyms</title>
            <author initials="D." surname="Chaum">
              <organization/>
            </author>
            <date year="1981"/>
          </front>
          <seriesInfo name="Communications of the ACM" value="24(2) 84-90"/>
        </reference>
        <reference anchor="RSW1996">
          <front>
            <title>Time-lock puzzles and timed-release crypto</title>
            <author initials="R. L." surname="Rivest">
              <organization/>
            </author>
            <author initials="A." surname="Shamir">
              <organization/>
            </author>
            <author initials="D. A." surname="Wagner">
              <organization/>
            </author>
            <date year="1996"/>
          </front>
          <seriesInfo name="MIT Technical Report" value="MIT/LCS/TR-684"/>
        </reference>
        <reference anchor="Wesolowski2018">
          <front>
            <title>Efficient Verifiable Delay Functions</title>
            <author initials="B." surname="Wesolowski">
              <organization/>
            </author>
            <date year="2018"/>
          </front>
          <seriesInfo name="IACR ePrint" value="2018/623"/>
        </reference>
        <reference anchor="BBBF2018">
          <front>
            <title>Verifiable Delay Functions</title>
            <author initials="D." surname="Boneh">
              <organization/>
            </author>
            <author initials="J." surname="Bonneau">
              <organization/>
            </author>
            <author initials="B." surname="Bunz">
              <organization/>
            </author>
            <author initials="B." surname="Fisch">
              <organization/>
            </author>
            <date year="2018"/>
          </front>
          <seriesInfo name="CRYPTO" value="2018"/>
        </reference>
        <reference anchor="ReschPlank2011">
          <front>
            <title>AONT-RS: Blending Security and Performance in Dispersed Storage Systems</title>
            <author initials="J. K." surname="Resch">
              <organization/>
            </author>
            <author initials="J. S." surname="Plank">
              <organization/>
            </author>
            <date year="2011"/>
          </front>
          <seriesInfo name="FAST" value="2011"/>
        </reference>
        <reference anchor="Loopix2017">
          <front>
            <title>The Loopix Anonymity System</title>
            <author initials="A. M." surname="Piotrowska">
              <organization/>
            </author>
            <author initials="J." surname="Hayes">
              <organization/>
            </author>
            <author initials="T." surname="Elahi">
              <organization/>
            </author>
            <author initials="S." surname="Meiser">
              <organization/>
            </author>
            <author initials="G." surname="Danezis">
              <organization/>
            </author>
            <date year="2017"/>
          </front>
          <seriesInfo name="USENIX Security" value="2017"/>
        </reference>
        <reference anchor="BB84">
          <front>
            <title>Quantum Cryptography: Public Key Distribution and Coin Tossing</title>
            <author initials="C. H." surname="Bennett">
              <organization/>
            </author>
            <author initials="G." surname="Brassard">
              <organization/>
            </author>
            <date year="1984"/>
          </front>
          <seriesInfo name="IEEE International Conference on Computers, Systems and Signal Processing" value="175-179"/>
        </reference>
        <reference anchor="SCITT-ARP">
          <front>
            <title>draft-hillier-scitt-arp: Attestation Resolution Profile for SCITT</title>
            <author initials="J. D." surname="Hillier">
              <organization/>
            </author>
            <date year="2026"/>
          </front>
        </reference>
        <reference anchor="E8VS">
          <front>
            <title>draft-hillier-certisyn-essential-eight-verified: Verified Profile for the Australian Essential Eight</title>
            <author initials="J. D." surname="Hillier">
              <organization/>
            </author>
            <date year="2026"/>
          </front>
        </reference>
        <reference anchor="AIGVS">
          <front>
            <title>draft-hillier-certisyn-ai-governance-verified: AI Governance Verification Scaffold</title>
            <author initials="J. D." surname="Hillier">
              <organization/>
            </author>
            <date year="2026"/>
          </front>
        </reference>
      </references>
    </references>
    <?line 1476?>

<section anchor="acknowledgments">
      <name>Acknowledgments</name>
      <t>The author thanks the broader cryptographic-research community for
the foundational work cited herein, including Shannon's 1949 proof
of OTP security, Shamir's 1979 secret sharing construction, Chaum's
1981 mixnet, the Rivest-Shamir-Wagner 1996 time-lock puzzle, the
Boneh-Bonneau-Bunz-Fisch 2018 VDF construction, the Loopix mixnet,
the Bennett-Brassard 1984 quantum key distribution protocol, and
the NIST Post-Quantum Cryptography project.</t>
      <t>The author also acknowledges the operational lessons of the Soviet
VENONA programme (pad reuse as failure mode), the Crypto AG /
Operation Rubicon programme (hardware supply chain as compromise
vector), and the public disclosures of the NSA TAO catalogue
(endpoint implant as residual attack surface), which together
motivated the operational hardening described herein.</t>
    </section>
  </back>
  <!-- ##markdown-source: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-->

</rfc>
