[unpaidmoon]

Change and mathematical scope

Audit, not a new κ. The selected conditional exponent stays 0.00310045 (#5599). Merged #5718 is not admitted or rejected.

The published #5719 checker accepts its pinned PR5684 word and returns PASS_FIXED_CAUSAL_LEDGER_AND_TYPED_CONTACTS_ONLY. On the merged #5718 bytes it returns BINDING_UNRESOLVED with scientific_falsifier false: the pinned 106,613-byte cap refuses the 120,522-byte successor before any scientific scan.

An independent census of the two authenticated words shows the certificate's premises do not transfer:

• Four anchors are absent, including both field-715 moves 3 → 598 and 598 → 396. The only move of field 715 off frame 3 is MOVE 715,3,396 at index 15,072.
• The five surviving anchor tuples are unique at displacements 91, 91, 89, 90, and 92, so no single index translation carries the certificate.
• The sole field-712 write in [823, 14983) is ADD 712,748,1 at frame 598, not the certified ADD 712,747,1.
• Index 911 reads field 715 as a source (ADD 206,715,2 at frame 3), which the observer exclusion forbids.
• The provenance window [11202, 14983) contains an extra ADD 340,344,1 at frame 70, index 11,260, besides the credited provenance ADD at index 11,293.
• The numeric stability interval [14654, 14983) is still empty of field-340 writes, but its endpoints are different operations. That boolean is not a transferred proof.
• Owned INIT/CLEAR frame sequences shift uniformly by one label. Role 747 on #5718 reproduces role 746 on the old word, not role 747.

The frozen #5717 candidate is a third word, SHA256 9fd67dd89ed24a644257cd9aef06934ed4c6fbe0441804684a968868f2ef66a8. This package does not rerun that adapter.

Task and overlap

Base de5ec2e7a6074068452f45b1a191b5c72832df4a, priority revision 8. Card supplier-boundary-audit, scope refined to the #5718 delta named in the card's first experiment.

Session r56, item 4 of the ordered eligible list. The 19 October handoffs are already delivered or closed. The revision-8 "Next bounded decisions" table then orders the open rows; catalogue cards after that table were not reached. Rows 1–3 and 5–6 stayed open and were left to the other sessions.

Nearest work is the input packages themselves. #5719 says in prose that the certificate does not transfer. #5718 is the successor word. Neither evaluates the old anchors on the new word. Overlap refresh at 2026-10-27T04:26:22Z covered 388 open PRs and 5,262 discussion records; the index-level facts (displacements, the read at 911, the write at 11,260) did not appear outside this package.

Validation

Python 3.12, standard library only:

python3 scripts/research_package.py --package research/p104-5718-causal-coverage-gap --check-only
python3 -B research/p104-5718-causal-coverage-gap/verify.py
python3.12 scripts/research_package.py --package research/p104-5718-causal-coverage-gap

The registered replay prints PASS_P104_5718_CAUSAL_COVERAGE_GAP and PASS_REGISTERED_RESEARCH_PACKAGE. Controls: the published checker must accept only its pinned word; a destination-only field-715 ADD stays allowed; an injected source read at index 900 is reported; changing the role-748 birth frame breaks the +1 lifetime match.

Not run: the #5718 scientific replay, the #5717 adapter, the #5599 caller replay, Lean, and Docker.

Credit

The fixed subject and checker are from Chris R / Clear1nsight, research/p104-temporal-fixed-independent-audit, following Ali Romman's PR5713 and Joel Pulikkan / GamingPuzzled's PR5684 word. The successor word is from flyingviper, research/p104-two-boundary-helper-split. The third candidate hash is from research/p104-caller-owned-interface. No file from those packages is modified.

Prepared by unpaidmoon with substantial Cursor Grok assistance. Apache-2.0.

Open next questions

1. Does a rebased ledger, using the direct field-715 move from frame 3 to frame 396 and classifying the extra field-340 write at index 11,260, hold on the #5718 word? First write a new subject from the anchors in the package and scan that word alone.
2. Does the published #5717 adapter accept the #5718 word under an explicit +1 label map? The frame sequences match that shift, and the packaged candidate hash does not. Its entry point currently authenticates a different word.

Workflow feedback (optional)

The revision-8 priority table names #5718 as the successor that #5717/#5719 do not automatically cover, which made the bounded gap easy to select. The published checker refuses other bytes before the scientific predicates, so the transfer question has to be reimplemented rather than rerun.