Sent — cross-fence originals

Documents here are authored in this house (UCCA) and are the originals of record for artefacts whose copies cross to another house (RTOpacks) under FENCE-PROTOCOL-01. This is the mirror of canon/received/: received copies are theirs, held unedited; sent originals are ours, kept authoritative.

Rules:

• Originals of record. The authoritative bytes live here; the copy that crossed the fence is the derivative. Each side files its own record of a shared ruling (per FENCE-PROTOCOL-01 §4) — e.g. UCCA-FENCE-RULING-RECORD-01 (ours) mirrors FENCE-RULING-RECORD-01 (theirs, received).
• Non-canon. These are crossing/record artefacts, not canon. Where a ruling in one hardens into canon, it does so through a canon doc (e.g. Ruling 2 → UCCA-INTAKE-01), not by promotion.
• Filed verbatim, hash-verified. Filed from the authoring session's bytes, checked against the drop manifest (shasum -c). Not retyped.
• A CARRY IS NOT COMPLETE UNTIL IT IS RECORDED HERE. Ruled by Tim 2026-08-01 (UCCA-RULING-BRIEF-STANDING-RULES-2026-08-01 rule 2, as amended). Moving a document to ## Carried is part of the carry, not bookkeeping after it. Why it is load-bearing: replace-in-place is permitted only where never crossed is verified from this file at the moment of replacement. This file is our record and is written after the event it records — so a row left stale reports "awaiting carry" for a document that has already crossed, and a replacement made on that reading breaks digest identity silently, producing no error, only two houses holding different bytes under one doc_id. The verification is only as good as this file's freshness.

Contents (filed 2026-07-01, manifest-verified)

doc_id	what it carries
UCCA-FENCE-RULING-RECORD-01	conformer/gate custody + the warranty boundary (Ruling 1, Ruling 2)
UCCA-ASSESSMENT-RTOPACKS-BRIEF-01	the engine's assessment of RTOpacks' engine brief (R-, D-, B-* refs)
UCCA-ASSESSMENT-RTOPACKS-BRIEF-01-ADDENDUM-01	the engine's answer to the brief's §5 (A-* refs)
UCCA-ACK-CORRECTION-01	acknowledgment of RTOPACKS-CORRECTION-01 (C-3, C-4 citation kinds)

Awaiting Tim's carry

doc_id	what it carries	status
UCCA-CROSSING-PROBE05-RESULT-RECEIPT-01	the receipt for RTOP-CROSSING-PROBE05-RESULT-01 (4d918154…, filed 9fd0e64) — §1–§4 under form; the A-3 extension accepted into the combined answers crossing at this house's clock, with the not-retained posture stated; their §5 receives nothing by its own terms	FILED 2026-08-03 — awaiting Tim's carry. NOT CARRIED. Filing precedes relay per VERIFY-BEFORE-CROSS (Class G). Verifies to 2a2a48fe337f…, 5,196 B, 36 lines; digest publishes WITH the carry. Bytes frozen at Tim's approval — re-verified on disk before filing and matching the frozen figure exactly; a frozen crossing that fails its own digest never files and never crosses. The carry is Tim's act after this push is reported, not before.

Register state as at 2026-08-03 (revised) — ONE artefact filed and awaiting carry: UCCA-CROSSING-PROBE05-RESULT-RECEIPT-01, frozen at Tim's approval, filed before relay, digest publishing with its carry. ⚠ This line has gone stale TWICE today and been caught twice — it read ZERO on 2026-08-02 while UCCA-CROSSING-PROBE-ANSWERS-01 sat filed with no row here, and it read ZERO again for the minutes between filing UCCA-CROSSING-CONTROL-RECEIPT-AND-SUBSTRATE-REATTEST-01 and updating the count. Re-date and re-count IN THE SAME COMMIT that moves any row; a stale awaiting line is what licenses replace-in-place on bytes another house already holds. Carried artefacts are listed in the dated Carried sections below. RTOpacks' FI-07 finding was that neither house makes a filed-and-uncarried artefact visible with its age; this line answers it.

Carried — 2026-08-03

doc_id	what it carries	status
UCCA-CROSSING-CONTROL-RECEIPT-AND-SUBSTRATE-REATTEST-01	the receipt for RTOP-CROSSING-CONTROL-RESULT-01 (00382156d5d3…) under form, with one leg defect named and its cure asked for; and the four substrate fingerprints re-attested live for the pending solo re-fire — gate, reasoner, container and D1 schema, all unchanged from the substrate that served the control job	CARRIED 2026-08-03 by Tim, digest published with the carry per §3.3 · RECEIPTED BYTE-EXACT. RTOpacks' consolidated carry block of 2026-08-03 (item 3) states received, verified byte-exact their side — a530bbd05447be3c34206c343dba07466cf5c2cc1ca60aab0d54a8ae2e6ec571, 4,069 B, 47 lines, their independent computation — and filed. ⚠ Provenance stated, not blurred: the confirmation travelled in the carry block, not yet as a filed receipt artefact; a receipt artefact is cited alongside this line if one later crosses. Filing preceded relay. Approved by Tim 2026-08-03 as drafted; bytes frozen at approval. Verifies to a530bbd05447…, 4,069 B, 47 lines; digest publishes WITH the carry per ratified §3.3. Approved by Tim 2026-08-03 as drafted, no amendments — bytes frozen at approval. ⚠ §1 names an FI-10 instance on their carry leg — no digest travelled with RTOP-CROSSING-CONTROL-RESULT-01; the figure we hold is our own computation, and the crossing asks for a one-line confirmation against their filed original as the cure. Our received row stays computed-here until that arrives. ✅ §2 carries the D1 fingerprint WITH ITS DERIVATION, and shows the counter-example — the same schema hashed over column names alone yields a different number. A figure without its method is not checkable, and this crossing proposes the method travel in both houses' future fingerprint exchanges. ✅ Records that their §3 count (226 attempts, completed ×217) matches this house's independent gate-side count of 226 exactly — two houses, two instruments, one number.

Carried — 2026-08-02

doc_id	what it carries	status
UCCA-CROSSING-PROBE-ANSWERS-01	the probe's answers on deployed bytes — the §4 bound live in v37; the diagnostic record verbatim; the counters NULL with the instrument defect owned as ours; the filed status set corrected three ways; and the second accepted job on the byte-identical payload stated as bytes with one direct question	FILED 2026-08-02 at 1ae6d42 (register row eaae9b4) · CARRIED 2026-08-02 under §3.3 · RECEIPTED BYTE-EXACT. Verifies to ecf8d74bdeb1…, 11,055 B, 150 lines. Receipt is not testimony: RTOP-CROSSING-PROBE-DECLARATION-CORRECTION-01 (5a4c6c446992…) records it "received under form, digest confirmed both sides" and quotes ecf8d74b…/11,055/150 — matching our filed object on all three measures, computed independently in the other house. ✅ Its §5 question is ANSWERED by that same correction: the undeclared job 0ac575ff… was their fire. ⚠ THIS ROW WAS MISSING UNTIL 2026-08-03. The document was filed with a register row and none here, so this file — the authoritative carry record — could not see it, and the dated line above read ZERO awaiting while it was untrue. That is the FI-10 failure mode, reproduced by the executing seat on the same day it wrote the FI-10 entry. Repaired here rather than quietly added.
UCCA-CROSSING-SUBMISSION-CREDENTIAL-RESPONSE-01	this house's answer to RTOP-CROSSING-SUBMISSION-CREDENTIAL-01 (ac51e47f…) — Ask 1 the credential, with two of this house's own record defects owned: the reissue that never crossed, and the two carries never recorded; Ask 2 the value travels out-of-band through Tim and never touches an artefact, terms travel here; Ask 3 the gate contract holds with two refinements they do not hold, and course_code does not exist at input — our own §5.9 withdrawn as a current-state claim; Ask 4 both instruments live, brief_in_prompt never a field, the ceiling carry unproven with a re-carry offered, total = 0 named as owed.	CARRIED 2026-08-02 · FILED 2026-08-02 at 2ae2579, IN THAT ORDER. Filing preceded relay. ✅ RECEIVED BYTE-EXACT, BOTH DIRECTIONS. Their §1 computes 82588ae182a7…, 15,301 B, 171 lines on the received bytes — identical to our filed object; our confirmation of their figures crosses in UCCA-CROSSING-CREDENTIAL-THREAD-CLOSE-01. ⚠ NO DIGEST WAS PUBLISHED WITH THE CARRY ON THE RELAY LEG — their receipt records it as a defect of the leg, not of the filing, since the status block prepared digest-with-carry and the leg dropped it. That is the FI-10 pattern recurring on the very next carry, after it was logged, and it is recorded here rather than absorbed. It cost nothing only because both houses computed independently. ⚠ FI-11: the engine-side card UCCA-CARD-CREDENTIAL-RESPONSE-FILE-AND-REMINT-2026-08-02 travelled on this same leg and should not have; RTOpacks filed it immutable and acted on nothing. REV 3, FINAL — revs 1–2 replaced in place, never filed, never crossed. Evidence base UCCA-REPORT-CREDENTIAL-READS-2026-08-02 (10f24b014867…).
UCCA-CROSSING-TIER1-QUESTIONS-ANSWER-01	the answer to RTOP-CROSSING-TIER1-RESPONSE-ACK-01 on Q1 through Q4 — course_code; the credential, rotated rather than merely named; validation on the forward path; and the two empty-outcome payloads	✅ CARRIED — PROVEN ON BYTES 2026-08-02. Both houses computed independently and agree on all three measures; confirmed at RTOP-CROSSING-CREDENTIAL-RESPONSE-RECEIPT-01 (bd386116ac7f…) §2 — their words: "Both carries are hereby proven on bytes." Re-derived at this seat before promotion. ⚠ THE CARRY WAS MADE WITH NO RECORD KEPT AND NO DIGEST PUBLISHED — logged as FI-10. This row read NOT CARRIED while RTOpacks already held the bytes; the proof does not undo the record defect. History follows: REV 2 FILED 2026-07-31 — awaiting Tim's carry. NOT CARRIED. Filing precedes relay. Verifies to 9d1c6a2ecbea…, 143 lines, 20,022 B. ⚠ REV 1 (089d951e9824…, 143 lines, 19,425 B) WAS FILED AND THEN REPLACED IN PLACE BEFORE CARRY, for ONE FALSE CAPABILITY SENTENCE in its verification block. It told RTOpacks that R2 and D1 were "outside the verifying seat's reach". The seat did not reach them — that was true — but it COULD have, through a second Cloudflare connector it had not enumerated, and did-not and could-not are different claims. A reader weighing figures single-sourced to one agent is entitled to know the second reader could have checked and did not. Replaced in place rather than superseded by a new doc_id because rev 1 NEVER CROSSED — it sat in this section, never under Carried, and no other house holds its bytes, so the digest-identity rule does not bind. Exactly two frontmatter lines changed, body byte-identical, verified by diff. Do not publish 089d951e… with the carry — publish 9d1c6a2e…. Its digest is to be published WITH the carry — per the convention ruled 2026-07-31, and unlike the tier-1 response, whose publication remains OPEN (see its row below). Home-side working: UCCA-FINDING-GATE-DEPLOYED-CONTRACT-2026-07-31 and UCCA-ADDENDUM-ANCHOR-ASYMMETRY-2026-07-31.
UCCA-ENGINE-RUNNABLE-STATE-2026-07-31	the engine's current runnable state, answering RTOP-ENGINE-STATE-REQUEST-2026-07-30 — what the engine takes, what it emits, how it is invoked and by whom, what it has actually been run on, what breaks, timing and cost. Carries one real UCCO envelope verbatim at §2.2 (1,939 B, sha256 50715e26dd48a542…, pasted as bytes and never through a JSON parser)	✅ CARRIED — PROVEN ON BYTES 2026-08-02. Both houses computed independently and agree on all three measures; confirmed at RTOP-CROSSING-CREDENTIAL-RESPONSE-RECEIPT-01 (bd386116ac7f…) §2 — their words: "Both carries are hereby proven on bytes." Re-derived at this seat before promotion. ⚠ THE CARRY WAS MADE WITH NO RECORD KEPT AND NO DIGEST PUBLISHED — logged as FI-10. This row read NOT CARRIED while RTOpacks already held the bytes; the proof does not undo the record defect. History follows: REV 6 FILED 2026-08-01 — awaiting Tim's carry. NOT CARRIED. Filing precedes relay. Verifies to 495148b7db0c6562…, 34,870 B. Not claimed to be the final revision — rev 5's card made that claim once and it did not survive contact. ✅ REV 6 corrects a COUNT, not a claim: §0.3 said structureFault() "names exactly two things"; it returns three field paths — job_type, triumvirate, triumvirate.outcomes. One sentence, diff extent 7c7 and 46c46, §2.2 envelope bit-for-bit unchanged. ✅ REV 5 IS AN ADDITION, NOT A CORRECTION — a boundary stated, not an error fixed. Nothing in revs 1–4 is withdrawn. RTOpacks' thin thread assumes a qualification is an input; the document previously answered only that the engine had never been given one — true, and the weaker truth. §0.3 now states the boundary positively: one job carries one obligation, a twelve-unit qualification is twelve jobs and twelve envelopes, and nothing on this side assembles them. Every claim in it re-derived from bytes before filing — the gate's structureFault(), the frozen v1 payload's single-string source_code, and zero occurrences of "qualification" in the gate or in reasoner/*.py (its only occurrences in that package are the aviation and pharma calibration probes, in the equipment sense). Three anchored edits at §0's heading, §0.1 and the new §0.3; diff extent 7c7, 22c22, 28c28, 40a41,46, 41a48,55; §2.2 envelope bit-for-bit unchanged. ✅ RESOLVED AT REV 6. The rev 5 flag is kept here rather than deleted, so the register carries both the flag and its closure: §0.3 said "names exactly two things" against three actual field paths. The substantive claim (no unit, no qualification, no course, no domain vocabulary) was exact throughout; only the count was wrong. The executing seat caught it, applied rev 5 verbatim as its card required rather than silently fixing it, and flagged it — and rev 6 corrected it before carry. Found after filing and before crossing, which is the order this house prefers. ⚠ REV 4 (b67701397f300adc…, 31,195 B) was an evidence update, not a defect correction — read it as the system working, not as a fourth mistake. A recon brief ran, UCCA-FINDINGS-TGA-SEPARATION-2026-08-01 (d581db4cb03e7945…, filed in the same commit) found something material, and it reached the document before the carry rather than after. CHCPRT025 carries three elements, and the 45 forward payloads resolve to four distinct triumvirates, not one: 31 jobs ran ONE element, 9 ran all three, 3 ran two, 2 ran none. "One unit, 64 times" was true of the code and wrong about the substance, and §6's timing and cost figures are weighted toward one-element runs — a bias running in this house's favour, in a document RTOpacks scope against. Four anchored edits at §0.1 (two), §4 and §6; diff extent 7c7, 28c28,30, 145a148,149, 213c217; §2.2 envelope bit-for-bit unchanged. ⚠ REV 3 (f6c3df37e1b8ef5c…, 29,253 B) — the prior filed revision. ⚠ REV 2 (c556309a7b6d9499…, 28,437 B) WAS ALSO FILED AND REPLACED IN PLACE BEFORE CARRY — it fixed rev 1's false sentence and introduced a DIFFERENT one in the same line. Rev 2's §10 item 3 said ONE artefact was filed and awaiting carry, while the register line below it — written into the same commit — says TWO, the second being this document itself. The document excluded itself from its own count. That is the slice defect this house has now logged four times in a week (7-vs-13, 5-vs-9, 34-vs-56, and this): a true statement about part of a set asserted of the set. Corrected to name both artefacts. Exactly two lines changed — the frontmatter status: and §10 item 3 — verified by diff (7c7, 294c294); body byte-identical, §2.2 envelope block bit-for-bit unchanged at 1,939 B / 50715e26dd48a542…. ⚠ REV 1 (6d9b9f3c8f3ec38d…, 28,065 B) WAS FILED AND THEN REPLACED IN PLACE BEFORE CARRY, for ONE FALSE SENTENCE at §10 item 3. It stated that UCCA-CROSSING-RPL-CAPABILITY-ANSWER-01 had been "filed and uncarried since 2026-07-06" — but this index records that artefact under ## Carried — 2026-07-31, so the document was telling RTOpacks something false about our own fence state, in the very section that governs crossing sequencing. Corrected to name the artefact actually awaiting carry. Replaced in place rather than superseded by a new doc_id because rev 1 NEVER CROSSED — it was filed 2026-08-01 and no other house holds its bytes, so the digest-identity rule does not bind. Replace-in-place has now run THREE times — UCCA-CROSSING-TIER1-QUESTIONS-ANSWER-01 rev 1, and this document's rev 1 and rev 2. It is a convention, not an exception, and its whole safety rests on never crossed being verified each time from this index rather than assumed. Publish ONLY the rev 6 digest with the carry — 495148b7db0c6562…. Not 6d9b9f3c…, c556309a…, f6c3df37…, b6770139… or df673499….
UCCA-CROSSING-FENCE-PROTOCOL-02-PROPOSAL-01	the FENCE-PROTOCOL-02 proposal — §3.3 disk-first inbound, §4 tracks, §5 thread concurrency, §6 split-confirmation form, §8 annex	FILED 2026-08-02 at 6752a1f4 · CARRIED 2026-08-02 (Tim-attested; digest published with the carry). Filing preceded relay. Verifies to 7566e2ef7f3d2563…, 15,253 B, 202 lines. Approved by Tim 2026-08-02: all five PROPOSED provisions as drafted, no amendments — bytes frozen at approval, checked on all three measures before filing and re-derived from the committed object after. ⚠ GOVERNANCE TRACK — on carry it became the one governance crossing in flight, per its own §4. Awaiting RTOpacks' seconding or counter; their receipt arrives under its own digest and is recorded on the received side when it does. NOT RATIFIED AND NOT MINTED: its §10 sets ratification (RTOpacks seconding + Tim) as the minting event, so no ruling-class register row exists for it.
UCCA-CROSSING-REVISION-PATH-SECOND-01	this house's second on the revision path — answers RTOP-CROSSING-REVISION-PATH-REPLY-01 (a9670112…), filed in received/. The corpus home seconded; the item-1 constraint accepted as a joint position — the type declaration must explain the 16/8↔3-module mechanism, not merely pick a type; item 2's negatives accepted verbatim; item 4's fixture halves both accepted and their catch owned. §6 carries one record note.	CARRIED 2026-08-02 — FILED 2026-08-02, IN THAT ORDER. Verifies to 14e37b5a79a3…, 10,586 B, 140 lines — derived from the committed object at c98bfd9, not from the filing card's report, and matching it exactly. Digest published with the carry, bytes frozen at approval. ⚠ FI-01-CLASS ORDERING DEFECT, RECORDED NOT HIDDEN: filing did NOT precede relay. It crossed before any copy existed in this repo, against VERIFY-BEFORE-CROSS (Class G). No byte or verification damage — unlike UCCA-CROSSING-AB-SCOPE-ACCEPT-01 (2026-08-01, the first of this class), the digest was published with the carry, so what crossed and what is filed here are provably the same artefact throughout the interval. Logged as FI-09 on Tim's ruling 2026-08-02. Different track from the governance crossing — the document's own frontmatter names UCCA-CROSSING-FENCE-PROTOCOL-02-PROPOSAL-01 (7566e2ef…) as "governance track, untouched here", so the single-file statement in the register line above is not disturbed. Nothing owed from this house in this thread — it closes our side; the next move is RTOpacks' only if they counter an acceptance.

Carried — 2026-08-01

doc_id	what it carries	status
UCCA-CROSSING-EXCHANGE-MODEL-RESPONSE-01	this house's response on the exchange model — answers RTOP-CROSSING-EXCHANGE-MODEL-POSITION-01 (92fc135070b2…), filed in received/	FILED 2026-08-01 · CARRIED 2026-08-01. Filing preceded relay — committed at c51c9de0 before the carry, which is the discipline FI-01 exists for and it held. Verifies to 970a4659b501…, 12,521 B, 170 lines. RTOpacks verified the RECEIVED bytes and reported exact on all three measures, final line echoed; their inbound digest for our receipt block was recomputed against their own filed original and was also exact. ✅ CAVEAT CLEARED 2026-08-02 — RESOLVED BY WITHDRAWAL, NOT BY RECEIPT. The awaited RTOP-RECEIPT-EXCHANGE-MODEL-RESPONSE-01 never existed as a carried artefact: RTOpacks filed it to their crossings/sent/ on 2026-08-01 and never carried it, and have withdrawn it as superseded by RTOP-CROSSING-REVISION-PATH-REPLY-01 (a9670112…). Authority: RTOP-NOTICE-RECEIPT-WITHDRAWN-01, fbbc51811ecd…, 2,276 B, 35 lines, filed in received/ — their words: "our reply's 'crossed separately' asserted a delivery that had not happened." Nothing further is owed inbound on this row. The acknowledgement of our bytes remains TM-attested (source of record UCCA-TM-2026-08-01-H §2.1) — the withdrawal disposes of the missing document, it does not retroactively make the acknowledgement artefact-proven.
UCCA-CROSSING-AB-SCOPE-ACCEPT-01	this house's acceptance of the A/B scope	FILED 2026-08-01 — AFTER IT WAS RELAYED. Verifies to 5eb6bef10e02…, 6,947 B, 96 lines. ⚠ FI-01 ORDERING DEFECT, RECORDED NOT HIDDEN: filing did NOT precede relay. This crossed to RTOpacks before any copy existed on this machine — a check on 2026-08-01 found it absent from canon/sent/, absent from the project root, untracked, and with zero commits in the whole repo history touching it. The bytes arrived here only when they were placed over the bridge afterwards. Filing-precedes-relay exists so that what crossed and what we hold are provably the same artefact; for the interval between the carry and this commit, this house could not have proved that about a document it had already sent.
UCCA-CROSSING-CEILING-FIGURE-AND-BRANCH-01	the ceiling measurement and the branch it selects	FILED 2026-08-01 — relayed. Verifies to e9a04b588597…, 4,872 B, 68 lines.
UCCA-FINDING-FORWARD-ENVELOPE-ENUMERATION-01	the forward envelope enumeration finding	FILED 2026-08-01 — CROSSED, and receipt confirmed BYTE-EXACT by RTOpacks (a2cff18e24de…, 6,639 B, 101 lines; their words: "exact match"). Their independently computed figure agrees with ours to the byte and the line — the digest-on-received-bytes convention working in both directions.

Carried — 2026-07-31

The carry was SPLIT, against Tim's one-parcel ruling — see fence-incidents.md FI-07. The engine-state response crossed first; the RPL answer followed separately the same day, on RTOpacks' asking. RTOpacks' receipt confirms both digests exactly — the first two-way byte confirmation across this fence, each house computing independently and then comparing.

doc_id	what it carries	status
UCCA-CROSSING-TIER1-RESPONSE-01	the answer to RTOP-CROSSING-TIER1-ANSWER-01 — their §5a answered on ADR-0005; the forward contract handed over corrected after two of this house's own claims were withdrawn rather than sent; the application domain-drift flag; and what submits the payload declared openly as NOT ESTABLISHED	FILED 2026-07-31 · CARRIED 2026-07-31. Filing preceded relay. Verified 31d6ebc6dade…, 140 lines, 17,434 B — and RTOpacks' independently computed digest matches exactly, quoted back at RTOP-CROSSING-TIER1-RESPONSE-ACK-01 §Receipt, which also records that they filed it before reading it. ⚠ Whether OUR sha256 was published with this carry is OPEN. An earlier wording of this row asserted that it was — present tense, drafted before the carry, never checked — and that assertion is withdrawn. Weak contrary evidence, recorded as inference and not as fact: RTOpacks' receipt states only "computed at this end", and that house is meticulous about naming a published digest when one arrives. If it did not go, the convention was breached on the first crossing it governed and this row is amended to say so.
UCCA-CROSSING-RPL-CAPABILITY-ANSWER-01	the engine-capability answer to RTOpacks' RPL query — evidence→competency reasoning is a new mode on the backward core, not a core re-engineer (bytes-hardened on the 2026-07-06 read-only diagnosis recon; ADR-0005 forward course-shape did not leak backward)	FILED 2026-07-06 · CARRIED 2026-07-31 — 25 days. Verified 721ad42dc6ca…, 134 lines, 8,666 B, matching RTOpacks' independently computed digest exactly. ⚠ Its own status: frontmatter field reads DRAFT — awaiting file-to-sent and is STALE IN THE BYTES, DELIBERATELY. It is a label, not a state; 721ad42d… is the filed copy. Corrected by record at UCCA-NOTE-RPL-STATUS-FIELD-2026-07-31 — read that note before the field. Never edited: never backdate, and RTOpacks now hold bytes that hash to this digest.
UCCA-CROSSING-ENGINE-STATE-RESPONSE-01	the answer to RTOP-ENGINE-STATE-REQUEST-2026-07-30 — yes to the runnable-state document, with the honest limit that most of what was asked is live substrate the drafting layer is blind to, so a recon pass with live reads comes first; names the convergence with UCCA's own Window 3; restates the non-negotiable (the engine raises hands and never signs, and no BETA label changes that); carries 23 questions in three tiers and authors nothing on RTOpacks' behalf	FILED 2026-07-30 UTC · CARRIED 2026-07-31. Filing preceded relay — it was not crossed at filing time, which is the discipline FI-01 exists for and it held. Verified 3a124d02f2f1…, 97 lines, 16,441 B, matching RTOpacks' independently computed digest exactly. It crossed first and alone, against the one-parcel ruling — FI-07.

No digest was published with the RPL or engine-state carries. Both were confirmed only because both houses computed independently and compared. For the tier-1 carry it is OPEN — see that row. The publish-the-digest-with-the-carry convention is ruled adopted in both directions — see the standing rule at the head of this file — and RTOpacks honoured it on their next inbound, RTOP-CROSSING-TIER1-RESPONSE-ACK-01, which is the first crossing in either direction to arrive checkable rather than confirmable by luck.

The contract pack these support (UCCA-CONTRACT-PACK-01) is filed on the technical docs surface at docs-site/docs/strategy/ucca-input-path-contract-pack-v0.1.md — it's an API/build spec, not a record.