Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
135 changes: 135 additions & 0 deletions research/p104-5718-causal-coverage-gap/README.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,135 @@
# The fixed #5719 causal certificate does not cover merged #5718

**Audit, not a new κ.** The selected conditional exponent stays
**0.00310045** (`#5599`). This package does not admit or reject the `#5718`
construction. It shows that one published coverage certificate does not apply
to that word.

## Result

Let `W0` be the authenticated PR5684 word used by
[PR #5719](https://github.com/CrocSwap/integer-mult-bounds/pull/5719), and let
`W1` be the authenticated word of merged
[PR #5718](https://github.com/CrocSwap/integer-mult-bounds/pull/5718).

The published `#5719` checker accepts `W0` and returns
`PASS_FIXED_CAUSAL_LEDGER_AND_TYPED_CONTACTS_ONLY`. On the bytes of `W1`,
renamed to the filename it requires, the same checker returns exit 3 and
`BINDING_UNRESOLVED`, with `scientific_falsifier` false and
`coverage_complete` false. The refusal is the pinned 106,613-byte cap. It is
not a scientific verdict on `#5718`.

An independent census of the same authenticated bytes then separates what
transfers from what does not.

| Premise named by the `#5719` certificate | On `W1` |
|---|---|
| Schema `p104-r22-reframe/1`, public count 743 | Schema `p104-terminal-main-split-short-cleanup/1`, public count 744, program length 25,546 |
| Strict frame chain, dimensions 0, 1, 3 at frames 3, 598, 396 | Same chain |
| Nine anchor events, each unique at its index | Four anchors occur nowhere. The other five occur once, at displacements 91, 91, 89, 90, and 92 |
| Sole field-712 write in `[823, 14983)` is `ADD 712,747,1` at frame 598 | The sole write there is `ADD 712,748,1` at frame 598, index 13,663 |
| Field-715 moves `3 → 598 → 396` | Those two moves are absent. The only move off frame 3 is `MOVE 715,3,396` at index 15,072 |
| No field-715 observer in `[823, 14983)` | Index 911 is `ADD 206,715,2` at frame 3 |
| One field-340 write in `[11202, 14983)`, the credited provenance `ADD` | Two writes: `ADD 340,344,1` at frame 70, index 11,260, and the credited `ADD` at index 11,293 |
| No field-340 write in the numeric interval `[14654, 14983)` | Still true, and not a transferred proof: index 14,654 and index 14,983 are different operations, the old read now lies inside the interval at 14,744, and the old cut lies at 15,075 |

The owned `INIT`/`CLEAR` frame sequences of roles 743 through 751 on `W0`
match roles 744 through 752 on `W1` exactly. Role 743 is unused on `W1`.
Role 747 on `W1` therefore reproduces role 746 on `W0`, not role 747. A
uniform label shift describes the lifetimes. It does not make the
certificate's field-747 claim true, and it does not restore the missing
715 chart path.

The frozen candidate of
[PR #5717](https://github.com/CrocSwap/integer-mult-bounds/pull/5717) is a
third word, SHA256
`9fd67dd89ed24a644257cd9aef06934ed4c6fbe0441804684a968868f2ef66a8`. Its
published binding names that hash. This package does not rerun the adapter,
so it does not confirm or refute the `#5718` author's report that the adapter
passed.

## Task and overlap

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

This is item 4 of the eligible list below. The nearest packages are the
inputs: `research/p104-temporal-fixed-independent-audit` states in prose that
its certificate does not transfer, and `research/p104-two-boundary-helper-split`
is the successor. Neither package evaluates the old anchors on the new word.
`research/p104-caller-owned-interface` authenticates a different candidate.

Ordered eligible list for session r56, rebuilt on 2026-10-27 against
`de5ec2e7a6074068452f45b1a191b5c72832df4a`. Open handoffs from the 19 October
index (discussion comments 19030045, 19030047, 19030049, and 19030053) come
first and are dropped: the `(w, q, v_N)` consumer, the fresh T385 wrapper,
the L546 six-tape bill, and the single-ADD retiming class are already
delivered by this account or closed on main. No newer finite handoff index
replaced that list. The priority table is the revision-8 "Next bounded
decisions" table in `docs/research/current-priorities.md`. A row stays while
its stated next decision is still open. Catalogue cards not named by that
table were not reached. Dropped before the cut: work this account already
delivered, results closed on current main, and topics already on a fork
`cursor/*` branch.

1. Complex helpers (`p104-side-correction`, `coverage-law-change`): one
uncovered helper group and causal cut with a complete caller bill.
2. BIT (`bit-helper-contract`): review the `#5679`/`#5708` chain before
another G20 search.
3. New decoders and interfaces: one complete changed producer/consumer
contract with an optimistic full signed bill.
4. **This package.** Supplier audit of merged `#5718` against the `#5719`
certificate and the `#5717` candidate hash.
5. All-size audit (`uniform-transfer-audit`): one selected assumption at its
actual caller.
6. Verification tooling (`replay-runtime`).

## Reproduce

Python 3.12, standard library only, from the repository root:

```sh
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
```

`verify.py` prints

```text
PASS_P104_5718_CAUSAL_COVERAGE_GAP absent_anchors=4 displacements=91,91,89,90,92 observer_read=911 owned_shift=1 direct_715_move=15072 checker_binding_refusal=true no_new_kappa=true
```

The checker hashes the five live inputs before reading them. It reruns the
published `#5719` program on `W0` and on a temporary copy of `W1`. The
injection control replaces one event of `W0` in memory: a destination-only
field-715 `ADD` stays allowed, and `ADD 206,715,2` at index 900 is the only
reported observer. A one-frame change to the shifted role-748 birth is
rejected by the lifetime comparison.

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

## 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 11260, hold
on `W1`? The old certificate cannot be that ledger. First write a new
subject from the anchors recorded here and scan `W1` alone. Nothing here
prices that scan or the `#5718` caller.
2. Does the published `#5717` adapter accept `W1` under an explicit `+1`
label map? The frame sequences match that shift, and the packaged
candidate hash does not. First point the adapter at `W1` and record the
map it uses. Its entry point currently authenticates a different word.

## 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.
60 changes: 60 additions & 0 deletions research/p104-5718-causal-coverage-gap/SOURCES.json
Original file line number Diff line number Diff line change
@@ -0,0 +1,60 @@
{
"schema_version": 1,
"base_commit": "de5ec2e7a6074068452f45b1a191b5c72832df4a",
"files": [
{
"path": "certificates/selected-result.json",
"usage": "provenance",
"git_blob": "581f8d87008354eb5b869f7ed9de1317ad837ce8",
"sha256": "2a0220e866417774606e0f991e41b9c1d26b1f43b97857238cf478dbc7d330b5"
},
{
"path": "docs/research/agent-task-catalog.json",
"usage": "provenance",
"git_blob": "3404fbdd5e7529a0674eef26114b5071e4586cec",
"sha256": "ead6cdc9d0af6d58e8a49744d8ec8c7c15e13d2ba93c276c434bcf9b51b93a81"
},
{
"path": "research/p104-caller-owned-interface/source/BINDING.json",
"usage": "input",
"git_blob": "d7152c152617961ef24d4b3301e85a620b21994f",
"sha256": "aff998b6d3069cea1a0a10b2a2fcd3b8100d3c84e521ef54b679a5c996839a99"
},
{
"path": "research/p104-caller-owned-interface/source/WORD.json.gz",
"usage": "input",
"git_blob": "45f6c10bf8a0ad6eb5f6714377a464e274db7dbd",
"sha256": "9fd67dd89ed24a644257cd9aef06934ed4c6fbe0441804684a968868f2ef66a8"
},
{
"path": "research/p104-temporal-fixed-independent-audit/inputs/LICENSE",
"usage": "input",
"git_blob": "261eeb9e9f8b2b4b0d119366dda99c6fd7d35c64",
"sha256": "c71d239df91726fc519c6eb72d318ec65820627232b2f796219e87dcf35d0ab4"
},
{
"path": "research/p104-temporal-fixed-independent-audit/inputs/NOTICE",
"usage": "input",
"git_blob": "fa0bd1401f5775f4351cfa23b08f3998dba6129e",
"sha256": "e8e0cabef6cf40d63f3cea3b6403dfd1af8727e40df03f09b2932718bcddc1b0"
},
{
"path": "research/p104-temporal-fixed-independent-audit/inputs/WORD.json.gz",
"usage": "input",
"git_blob": "f3fb0e2144ec8a5c3407e4eaa6528bb3436ba573",
"sha256": "d6db1d0d97b13ea483a462c65437664d15d34dec56c3682cf6e1de31266ec6c2"
},
{
"path": "research/p104-temporal-fixed-independent-audit/verify.py",
"usage": "input",
"git_blob": "a2111649f30f57f7acda419c678e950755e85258",
"sha256": "bce6cdf82cb1492d209c241670df894548a485b89622a39851f26352f61e999b"
},
{
"path": "research/p104-two-boundary-helper-split/WORD.json.gz",
"usage": "input",
"git_blob": "4db015aca34ea957dd8b3b4d7ae4fa12b89934f8",
"sha256": "cff7196f8b168c4ebf8227185eed98387b66ec5db0179108adfea4045775cc31"
}
]
}
97 changes: 97 additions & 0 deletions research/p104-5718-causal-coverage-gap/TASK.json
Original file line number Diff line number Diff line change
@@ -0,0 +1,97 @@
{
"schema_version": 1,
"task_id": "supplier-boundary-audit",
"priority_revision": 8,
"base_commit": "de5ec2e7a6074068452f45b1a191b5c72832df4a",
"original_card": {
"id": "supplier-boundary-audit",
"question": "Does one changed supplier have complete source-bound native, caller, ordinary and outer admission coverage?",
"starting_points": [
"docs/research/current-priorities.md",
"docs/research/research-family-status.md",
"certificates/selected-result.json",
"docs/research/pr-intake-evidence/pr5599-completion/PROOF-REVIEW.md"
],
"methods": [
"Construct and independently check one source-bound instance of the stated first experiment.",
"Seek an exact counterexample or target-specific complete-budget obstruction to that same proposed change."
],
"acceptance_gate": "Pin the current sources and one obligation; the historical owner audit is an example, not evidence of a current bug.",
"first_step": {
"command": [
"python3",
"-B",
"scripts/research_selected_baseline.py"
],
"expected": "PASS_RESEARCH_BASELINE_BINDINGS_ONLY",
"inspect": "Read revision 8, the selected review and live exact-source handoffs. Then follow this card's bounded first_experiment.",
"scope": "Source-bound orientation only, not a supplier replay or theorem proof."
},
"first_experiment": "Choose one latest uncovered candidate and build a delta coverage map against the selected review. For P104, #5717 changes owned-lifetime assumptions and #5719 covers exact #5713 causal premises; neither automatically transfers to #5718. For BIT, #5679/#5708 need component admission and compatible outer use. Check controls and exact source identity before reuse; stop with one resolved gap or explicit remaining obligation. A heavy full replay requires a separate serial resource budget.",
"bottleneck_link": "Judge the actual current supplier contract and complete signed target/paid bill; frozen-profile ceilings require an applicability proof.",
"allowed_variations": [
"One exact source head, helper/cut region or quantified proposition and changed hypothesis; check live handoffs first. Different seeds are not different scopes."
],
"scope_change_triggers": [
"Changing source head, helper/cut domain, operation alphabet or caller boundary without recording the new scope.",
"Treating an archived proposal or local numerical pass as complete supplier admission."
],
"overlap_terms": [
"supplier admission",
"5717",
"5719",
"5708",
"owned lifetime"
],
"existing_checks": [
{
"scope": "PR #4088: author-checked local words, native receipts, owner/phase fill and exact moment.",
"evidence": "research/staged544-mixed-fourth-entry/README.md",
"remaining": "Rebind and validate ordinary/outer admission with all four providers; full mixed release not established.",
"status": "author-checked",
"source_commit": "72d7efc181edb21806a878f174847c96dd410ab0",
"selected_record_sha256": "bb2eaa58647b353f88d0aec2eafb9d30df1be53589231b23e12ceef3e4bad129"
},
{
"scope": "PR #3801: successor scatter-boundary decision package.",
"evidence": "research/scatter-boundary-decision/README.md",
"remaining": "Consult its exact acceptance scope before any new exporter audit; not proof of a selected failure.",
"status": "author-checked",
"source_commit": "911f8800ba7a4b03603ef62c5ffc6240e53cb5fd",
"selected_record_sha256": "bb2eaa58647b353f88d0aec2eafb9d30df1be53589231b23e12ceef3e4bad129"
},
{
"scope": "Selected PR5599: complete helper-subspace construction and independent caller replay; five framework assumptions remain.",
"evidence": "docs/research/pr-intake-evidence/pr5599-completion/PROOF-REVIEW.md",
"remaining": "The next experiment must identify uncovered source-specific work. Successor archive intake does not transfer this admission.",
"status": "reviewed",
"source_commit": "09dd0965711ca7eeb169057a420abab1c9052137",
"selected_record_sha256": "2a0220e866417774606e0f991e41b9c1d26b1f43b97857238cf478dbc7d330b5"
}
]
},
"scope_status": "refined",
"actual_scope": "Does the fixed #5719 causal/typed-contact certificate, or the #5717 owned-label names, cover merged #5718?",
"scope_change_reason": "The card's first experiment asks for one latest uncovered candidate and one resolved coverage gap. This package answers that for merged #5718 against the #5719 certificate and the #5717 frozen candidate hash. It does not audit BIT #5679/#5708, rerun #5599 admission, or transfer those checks onto a later kappa proposal.",
"bottleneck_link": "A higher conditional P104 exponent can be selected only after the successor's own caller, causal, and owned-lifetime premises are covered. Showing that the #5719 certificate does not apply to #5718 blocks that particular reuse; it does not itself raise kappa.",
"budget_minutes": 60,
"stop_rule": "Stop when one source-bound gap is machine-checked: the published #5719 checker gives no scientific verdict on the #5718 bytes, and the fixed anchors, 715 chart path, provenance window, and role-747 lifetime do not hold on that word. Do not launch the #5718 or #5599 scientific replays.",
"outcome": "completed",
"evidence": "verify.py prints PASS_P104_5718_CAUSAL_COVERAGE_GAP. The predecessor checker accepts the pinned #5684 word and returns BINDING_UNRESOLVED, scientific_falsifier false, on the #5718 bytes. Independent census records four absent anchors, displacements 91,91,89,90,92, one forbidden 715 source read at index 911, and a uniform +1 owned-lifetime frame shift that makes role 747 a different lifetime.",
"author": "unpaidmoon",
"ai_assistance": "Substantial Cursor Grok assistance.",
"open_questions": [
{
"question": "Does a rebased ledger, using the direct 715 move from frame 3 to frame 396 and classifying the extra field-340 write at index 11260, hold on the #5718 word?",
"why_it_matters": "The missing 3-to-598-to-396 path and the extra provenance write are the premises a transferred certificate would have to replace. A rebased pass would be new coverage; this package only shows the old certificate does not apply.",
"first_step": "Write a new fixed subject for the #5718 anchors observed here and scan that word alone, without reusing the #5719 index literals.",
"blocked_by": "No published subject pins those replacement events. The #5718 author replay is a different checker and was not rerun."
},
{
"question": "Does the published #5717 owned-lifetime adapter accept the #5718 word under an explicit +1 label map?",
"why_it_matters": "The #5718 pull request says that adapter passed. Its published entry point is bound to a third word hash, so the claim is not the certificate executed here.",
"first_step": "Point the #5717 adapter at research/p104-two-boundary-helper-split/WORD.json.gz and record the label map it uses.",
"blocked_by": "The adapter authenticates source/WORD.json.gz inside its own package. Rebinding it is a separate source change, not a corollary of the frame-sequence shift."
}
]
}
26 changes: 26 additions & 0 deletions research/p104-5718-causal-coverage-gap/VERIFY.json
Original file line number Diff line number Diff line change
@@ -0,0 +1,26 @@
{
"schema_version": 1,
"status": "ready",
"python": "3.12",
"requirements": [],
"inputs": [
"research/p104-caller-owned-interface/source/BINDING.json",
"research/p104-caller-owned-interface/source/WORD.json.gz",
"research/p104-temporal-fixed-independent-audit/inputs/LICENSE",
"research/p104-temporal-fixed-independent-audit/inputs/NOTICE",
"research/p104-temporal-fixed-independent-audit/inputs/WORD.json.gz",
"research/p104-temporal-fixed-independent-audit/verify.py",
"research/p104-two-boundary-helper-split/WORD.json.gz"
],
"commands": [
{
"argv": [
"{python}",
"-B",
"{package}/verify.py"
],
"timeout_seconds": 120,
"checks": "Hash-pins both words and the published checkers, requires the #5719 checker to accept only its own word, and checks the anchor, provenance, chart-path, and owned-label delta plus injection controls."
}
]
}
Loading
Loading