Honest status (matches the live popup chip): OPEN · CANDIDATE (testable). Direction: held. STATUS-UPGRADES:0. Binding label: Conditional / Diagnostic; EXCLUDED cosmology sector (Axiom I / SG-10 §9.3.2). Frozen 13D K₆ branch
dcc66f1b2685/ manifesta5b1e6f9d951— READ-ONLY.
Headline. The frozen 13D geometry, run forward into the early universe, hands us a single surviving inflaton candidate — the breathing modulus σ rolling down an exponential plateau — whose scalar tilt lands dead-center on Planck as a consistency-check, and whose gravitational-wave signal is a pre-registered bet that LiteBIRD will settle by ~2030. The geometry does two pieces of genuine, target-blind narrowing work: it kills one of the two candidate inflatons by computation (the ρ mode is too stiff to inflate) and it fixes the tilt direction with no dial (the plateau slope is read off the dimension split). Everything past those two wins is honestly open, the spectrum's magnitude relocates onto un-traced conventions rather than reducing, and — decisively — the whole gate lives inside a sector the programme has declared out of scope.
The honest grade, in one paragraph. Gap-08 is OPEN (cascade), and it can never close in-scope. Cosmology — inflation, reheating, CMB, structure — is a declared excluded sector (Axiom I / SG-10 §9.3.2); a script-enforced lint forbids using this gate to support any in-scope claim, so even a perfect n_s/r match would remain a diagnostic consistency, never a derivation. Within that diagnostic lane the gate is also open on its own terms: the σ number is convention-anchored (it relocates onto at least two un-traced local conventions), three σ countersigns are uncleared, the loop coefficient c_loop is inherited-open, the absolute amplitude A_s is BLOCKED, and the spectrum's life-or-death is reserved for a future measurement. The honest ceiling is "a serious candidate consistency, pre-registered to die — NOT a validated spectrum, NOT an in-scope result."
What this dossier establishes, and what it does not. It establishes — at the grades the corpus records, and no higher — that: (i) the ρ-inflaton is a computed DEAD verdict at qualitative grade (KK-stiff); (ii) the surviving σ-inflaton is an exponential-plateau candidate whose plateau slope λ² = 4/K_σσ = 1/6 is dial-free given the frozen split n = (6, 2, 1); (iii) the projected scalar tilt n_s ∈ [0.9643, 0.9679] sits Planck-dead-center as a consistency-check, with a red-tilt direction that is near-universal across the four named normalization branches; (iv) a convention-independent, pre-registered union falsifier exists: r outside [3.5, 36]×10⁻³ kills the candidate across all four named conventions, and the current bound r < 0.036 is survived; and (v) the gravitational-coupling anchor κ₁₃² = 1/(π⁴ M_U⁹ M_Pl²) is closed-by-computation. It does not establish that λ² = 1/6 is a banked, derived output (it is a theorem-candidate, fail-closed on an unread frozen source); it does not claim n_s or r as predictions in the reserved sense; it does not claim a certificate-grade ρ-death; it does not claim an absolute amplitude; and it does not — and cannot — claim the gate closed. A given spectrum is not a derived spectrum; anchored ≠ derived; AXIOM-CLOSED ≠ proven; selection ≠ derivation; dissolved ≠ solved.
Status, spine, and every grade in this section are reproduced from …/TOE/PER_GATE_DOSSIERS/GAP08_COMPLETION_HANDOFF/01_DOSSIER.md (§0, §6), …/02_CURRENT_STATE.md, …/specialist_reply.txt (§1, §5), …/TOE/TOE_Self_close_chapters/ch_09_gap08_inflation.md (§09.0), and the regrade entry …/TOE/GATE_REGRADE_BESTCASE_AND_HOLES_2026-06-29.md (Gap-08, lines 1364–1414).
Modern cosmology has no first-principles answer for why inflation produced the spectrum we actually observe. The inflationary paradigm successfully explains the large-scale homogeneity, flatness, and near-scale-invariance of the cosmic microwave background (CMB) by positing a phase of quasi-de Sitter expansion driven by a scalar "inflaton" field rolling down a potential V(φ). But the paradigm is a framework, not a theory of a specific field: the three quantities a working cosmologist cares about —
together with the identity of the field that did the inflating and the number of e-folds N_ between the pivot scale's horizon-crossing and the end of inflation — are all fit to data, not derived. Hundreds of inflationary models exist; they are discriminated by data, not selected by first principles. No accepted theory predicts (n_s, r, A_s) from a fixed, independently-motivated geometry, and the single most decisive discriminator — a measurement of the primordial tensor amplitude r via CMB B-mode polarization — has not yet been made.*
This is the gap Gap-08 addresses: can a fixed, frozen higher-dimensional geometry — chosen for entirely non-cosmological reasons — supply an inflaton and a falsifiable spectrum, target-blind?
The measured comparators (held throughout as comparison targets, never inputs — see §7) are, to the precision the corpus uses:
| Comparator | Measured value | Role |
|---|---|---|
| Scalar tilt n_s | ≈ 0.965 (Planck-measured, "dead-center") | comparator |
| Tensor-to-scalar ratio r | r < 0.036 (current upper bound, MEASURED) | comparator / falsifier edge |
| Scalar amplitude A_s | ≈ 2.1×10⁻⁹ (Planck-measured) | comparator (for the BLOCKED A_s leg) |
| Effective neutrino number N_eff | 3.044 (in-band) | comparator (reheating interface) |
| Forthcoming tensor measurement | LiteBIRD, ~2030 | the decisive external adjudicator |
The headline experimental fact is the absence of a tensor detection: the current bound r < 0.036 rules out the simplest large-field (e.g. φ², "natural") inflation models but is consistent with a broad class of plateau models (Starobinsky/R², α-attractors, and the construction here), which generically predict small r. The forthcoming LiteBIRD satellite is designed precisely to reach the σ(r) sensitivity at which these plateau models live (roughly r ~ 10⁻³), making ~2030 the decisive moment.
Comparator values and the r < 0.036 bound: GATE_BRIEF_GAP08.md (Current honest status), ch_09_gap08_inflation.md §09.1, §09.5 (A_s ≈ 2.1×10⁻⁹), §09.14 (N_eff = 3.044).
The unfilled hole this gate targets. None of these derives the specific plateau slope from an independently-fixed geometry while staying target-blind, and none pre-commits a convention-independent tensor falsifier. The two genuine moves Gap-08 banks — a computed kill of one candidate inflaton, and a dial-free slope read off a frozen split — are exactly the moves the prior attempts cannot make. The honest counter-weight, stated up front: even those moves sit inside an excluded sector and the magnitude still relocates onto conventions, so this is narrowing, not closure.
Full construction lives in the published manuscript: Scoped TOE main statement §6.2 "Inflation branches and K13-2 (Gap 08)" + §7 truth-table, and Paper IV formal register (K13-2 source anchor, corpus line L10268). This section reproduces the attack-grade derivation and every banked number, each cited to the corpus file it was read from.
The geometry touches cosmology through one forward chain — no new geometry is searched, and no in-scope claim may rest on it:
frozen 13D split n = (6, 2, 1)
│
├─► two candidate inflaton directions {ρ, σ}
│ ρ ── KK-stiffness test ──► DEAD (qualitative grade; computed verdict)
│ σ ── breathing modulus rolls a potential ──► exponential plateau (v10 regime)
│
├─► plateau slope λ² = 4/K_σσ = 1/6 [read off n=(6,2,1); NO dial] — geometry-fixed (CANDIDATE)
│
├─► + K13-2 normalization conventions (III-a parity bit; III-b (9/11)² factor) + countersigns + c_loop
│
├─► PROJECTED n_s ∈ [0.9643, 0.9679], r ∈ [3.5, 10]×10⁻³ (operative branch T6)
│ ── compare (NEVER fit) against ──► MEASURED comparators (Planck n_s, r-bound, N_eff)
│
└─► union falsifier ──► LiteBIRD r outside [3.5, 36]×10⁻³ (all four branches) ⇒ candidate dies
Pipeline reproduced from 01_DOSSIER.md §1.1; ch_09_gap08_inflation.md §09.0, §09.18.
The surviving candidate is the geometry's breathing modulus σ — the scalar that controls the overall scale of the compact internal space. Its potential, assembled from the dimensional reduction, is (corpus §II):
V(σ) = c_KK e^{−4σ} + c_bdry e^{−2σ} + c_Wilson cos θ_W e^{−4σ} + c_loop e^{−6σ}
Each term has a geometric origin: the e^{−4σ} terms from Kaluza–Klein curvature and Wilson-line contributions, the e^{−2σ} boundary term, and the e^{−6σ} loop term whose coefficient c_loop is the one genuinely-computational residue (a 2-loop-equivalent FRG quantity; see §3.7 and §6). In the corrected (v10) regime the potential supports an exponential plateau, the regime that gives small r and the near-universal tilt.
Potential form and the four terms: C_c_loop_coefficient.md (what_is_known); ch_09_gap08_inflation.md §09.2.
A v10 Amendment records an important sub-result: with c_bdry negative across its band, the well is unconditional in c_loop — c_loop cannot threaten the existence of the well (the Gap-04 sign/bound result). Consequently, for Gap-08, c_loop's role is confined to the absolute amplitude A_s; it does NOT enter n_s, r, or the union falsifier. (This is the single most important scoping fact about c_loop — see §6, Hole D.) Source: C_c_loop_coefficient.md (v10 Amendment); ch_09_gap08_inflation.md §09.2(ii).
Of the two candidate inflaton directions the frozen split offers, the ρ ("breathing"/radial-type) direction is too stiff to inflate: it is KK-stiff, meaning its effective mass sits at the Kaluza–Klein scale, far above the inflationary (slow-roll) scale, so it cannot support a slow roll. This is a computed verdict, not an assertion — real narrowing of the candidate field from two to one.
Honest grade (carry verbatim). The ρ-death is at qualitative grade. A certificate-grade kill (an explicit quantitative KK-stiffness bound above the inflationary scale) is the optional sharpening — explicitly NOT required and NOT claimed; the σ-survivor carries the gap, so ρ does not need a sharper kill. The countersign slot that would formalize the verdict is pending. Separately, the NT-1 χ-leg run gate-halted on a scale artifact at v14, VOID both directions — "VOID both directions" means the run is inconclusive, not evidence either way; the ρ-leg qualitative-stiff verdict stands independently of it, and no value is asserted for the χ-leg.
ρ-DEAD grade, the optional certificate, the NT-1/χ VOID: ch_09_gap08_inflation.md §09.8, §09.9; 01_DOSSIER.md §2.1 (W3), §2.3(f); specialist_reply.txt R2.
This is the one number that would be a genuine geometric output. For the σ exponential plateau, the slope of the plateau is
λ² = 4 / K_σσ
where K_σσ is the σ-kinetic-metric coefficient. The claim is that the frozen dimension split n = (6, 2, 1) fixes K_σσ = 24, hence
λ² = 4/24 = 1/6
with no dial available to place it. The tilt direction — a red tilt (n_s < 1) — is geometry-fixed given the split, not fitted. This is the load-bearing structural datum of the gate.
The honest layering (this is where the rigor must be exact). The corpus checks this at three layers and fails closed at the third:
This fail-close is registered as BLOCKER A1-GEOM (see §6, Hole A). A distinct-quantity caution is recorded as a fail-safe: the K13-2 normalization datum K = 99/11 = 9 is the kinetic normalization constant — not the same symbol as K_σσ = 24 — and if they are the same object, the factor 24/9 = 8/3 needs an account (an OPEN consistency question, TOE-GAP-08-Q01).
Three-layer check, BLOCKER A1-GEOM, the 8/3 caution, the order-of-magnitude n_s ≈ 0.9636: ch_09_gap08_inflation.md §09.11; arithmetic re-verified in specialist_reply.txt §4 (line 149–150: "4/24 = 1/6, 4/9 ≠ 1/6, 24/9 = 8/3").
The σ exponential plateau (α-attractor-like family), evaluated in slow roll, gives a PROJECTED scalar spectral index
n_s ∈ [0.9643, 0.9679] (Planck-dead-center)
For an exponential plateau the tilt is dominated by the −2/N_ term and is only weakly sensitive to λ², which is exactly why the corpus records the tilt as "near-universal across branches" — robust to the T/D normalization and 6/u breathing convention choices, unlike r. The band's adjudication is by Planck, which is already MEASURED and dead-center; this comparison is a consistency-check, NOT a prediction in the reserved sense* (the Planck comparator is held strictly as a target).
An important reproducibility gap: the §IV inflation stanza publishes the n_s and r bands but does not publish the e-fold number N_ used to evaluate them — N_ is NOT STATED in the read corpus stanza, and the band's full reproducibility is therefore incomplete (see §6, Hole E). N_ is set by the reheating history, i.e. by Gap-09 (the reheating temperature band T_RH ∈ [2.4×10¹², 4×10¹⁵] GeV maps to an N_ window).
n_s band, near-universal tilt, consistency-check framing, N_ not stated:* ch_09_gap08_inflation.md §09.12, §09.10; 01_DOSSIER.md §0.1.
The tensor-to-scalar ratio r is where the K13-2 normalization ambiguity becomes experimentally adjudicable. The geometry produces a finite four-branch menu of r-bands, indexed by two binary convention axes — breathing (K₆-only "6" vs uniform "u") and normalization (textbook "T" vs D.1 "D"):
| Branch | Convention | r band (PROJECTED) | n_s | Standing |
|---|---|---|---|---|
| T6 | Textbook norm · K₆-only breathing | [3.5, 10]×10⁻³ | [0.9643, 0.9679] | FROZEN operative branch (v10/v11); slot PENDING |
| T_u | Textbook norm · uniform breathing | [6, 17]×10⁻³ (v15: [6.1, 17]) | Planck-centered | where a TEXTBOOK ruling lands if uniform breathing is operative |
| D6 | D.1 norm ((D−2)=11) · K₆-only | [8, 24]×10⁻³ | robust | survives r < 0.036 |
| D_u | D.1 norm · uniform breathing | [10, 36]×10⁻³ | robust | survives r < 0.036 |
The pre-registered, convention-independent union falsifier (carry verbatim):
If LiteBIRD measures r outside [3.5, 36]×10⁻³, the σ-inflaton is falsified regardless of [the four named] conventions.
The lower edge 3.5×10⁻³ is T6's floor; the upper edge 36×10⁻³ is D_u's ceiling. Inside the union, the four bands separate at LiteBIRD resolution — "the convention becomes an observable" (v11): the ~2030 measurement simultaneously tests the model AND reads off which of the four named normalizations nature uses. The current bound r < 0.036 (MEASURED) is survived by all four branches. The operative-branch (T6) forecast is recorded as "LiteBIRD-testable at 3.5–10σ" (PROJECTED).
The critical honesty caveat (carry verbatim). The falsifier POSIT is atomic, but the specific numeric band edges are PROVISIONAL-on-III-b. The convention-independence is banked only across the four NAMED branches (T6/T_u/D6/D_u); there is an additional un-traced freedom — the (9/11)² = 81/121 normalization factor (III-b, see §3.8) — that lies outside those four branches and is mentioned by neither convention axis. Until that factor is source-traced, the band edges inherit its unresolved freedom. So the falsifier must be stated honestly as: atomic posit; numeric band provisional. It is NOT "byte-fixed regardless of ANY internal bookkeeping choice."
Four-branch table, union [3.5,36]×10⁻³, T6 operative [3.5,10]×10⁻³, "convention becomes an observable," r < 0.036 survived: ch_09_gap08_inflation.md §09.13, §09.1; 01_DOSSIER.md §1.1. PROVISIONAL-on-III-b caveat: specialist_reply.txt R8 (lines 88–91), §3 (III-b).
The absolute scalar amplitude A_s rides two inputs and is currently BLOCKED. The amplitude relation is anchored by the 13D gravitational coupling, which is closed-by-computation (v11):
κ₁₃² = 1/(π⁴ M_U⁹ M_Pl²) — closed form, corpus-verbatim, at the frozen radii.
This is a genuine sub-win: pure geometry, no dial; the fold contributes a factor-2 band. But the absolute A_s value remains BLOCKED on two named inputs: (i) K13-2 — which operative branch/convention, fixed by the σ-map yaml read (B3); and (ii) c_loop — the σ⁻⁶ loop coefficient (inherited from Gap-04). The corpus standing non-claim is verbatim: "No absolute amplitude is claimed in this document." A tilt + ratio with no amplitude is not a spectrum. When B3 + c_loop discharge, the frozen A_s relation is evaluated and consistency-checked against Planck's A_s ≈ 2.1×10⁻⁹ — held as a comparison target only, with a ±factor-2 fold tolerance from κ₁₃ (so no per-entry fitting leaks in; a mismatch beyond that band is a recorded tension, not a silent re-tune).
κ₁₃ closed form, A_s BLOCKED on c_loop + yaml, the factor-2 fold band, Planck A_s ≈ 2.1×10⁻⁹ as comparator: ch_09_gap08_inflation.md §09.17, §09.5; specialist_reply.txt R4, R5.
The K13-2 convention is the σ-kinetic normalization, and it is the gate-defining residual. The early framing treated it as a single "one bit" / "one convention" choice. The specialist deliverable refutes that single-bit framing: R1 resolves into TWO distinct un-traced local freedoms — the σ number does not reduce, it RELOCATES:
III-a — the K13-2 parity bit. The σ-kinetic denominator: D.1 uses (D−2) = 11; the textbook 4D reduction would use (D₄−2) = 2. This axis is OWNER-RULED (retained as D.1, intentional, recorded 2026-06-15) and is external-to-AI: an AI cannot manufacture authorial intent the source does not record; attempting to "resolve" it would be fabrication. Reduced to Axiom at OWNER-DECISION grade, value-free, with a demotion clause retained. The support cited (NOT proof): the explicit D.1 occurrence at L10268, and that the D.1 denominator yields the integer normalization K = 99/11 = 9 (the textbook denominator would give the non-integer 99/2 = 49.5). This integer-K coincidence is support, not proof, and is NOT banked as a derivation.
III-b — the (9/11)² = 81/121 normalization factor. This is a separate, newly-exposed freedom. The arithmetic spine, re-verified exact: the recorded factor 81/22 = (11/2)·(9/11)² = 9²/(2·11). The (11/2) is the raw denominator ratio (D−2)/(D₄−2) = 11/2; the additional (9/11)² = 81/121 factor is mentioned by neither the breathing axis (B1) nor the normalization axis (B2). A corpus-wide search for "81/121" and "(9/11)²" in this context returns nothing (only an unrelated M_ scale-floor exponent), so III-b is genuinely un-source-traced. It is carried as a NEW named value-free posit, AXIOM-K13-2-RESIDUAL-NORMALIZATION-FACTOR, OPEN byte-trace, REDUCE_FURTHER: it demotes to a derived consequence of the SHAPE-SPLIT only if* an external read shows it follows from n = (6, 2, 1) target-blind.
Net. The gate carries ≥ 2 local conventions, not one — it RELOCATES, it does not reduce. The four convention branches span r across [3.5, 36]×10⁻³ — a factor of ~10 — so the magnitude is convention-anchored, not formula-forced. The breathing axis itself (III-a's partner) IS source-resolved: it is CLOSED by source (D.1.0 uniform-breathing declaration — the operative breathing mode is uniform across n = (6, 2, 1)).
Single-bit framing refuted, III-a parity bit, III-b (9/11)²=81/121 newly exposed, 81/22 = (11/2)·(9/11)², K = 99/11 = 9 vs 99/2, breathing axis CLOSED-by-source: specialist_reply.txt R1 (lines 30–47), §3, §4 (lines 148–153); ch_09_gap08_inflation.md §09.6, §09.16; regrade …2026-06-29.md lines 1384–1387.
For completeness, the corpus assigns 18 inventory rows to Gap-08, organized by convergence class. The actionable computational/mechanical lane (OPEN_TARGET) holds: c_loop (the FRG-2 number), the σ-map yaml read (B3), the absolute A_s (D2), the slope A1, the tilt A2, the r per-branch integration A3, N_* (A4), and the Gap-09/Gap-12 interfaces. The dominant terminal is the LiteBIRD measurement (OBSERVATION_BLOCKED) and the excluded-sector scope wall. Already-banked closures (reproduced, not new): the breathing axis (CLOSED-by-source) and κ₁₃ (CLOSED-by-computation). Optional sharpenings (not blockers): the certificate-grade ρ-death and the NT-1/χ re-run.
Full 18-row table and convergence-class legend: ch_09_gap08_inflation.md §09.0 (the inventory table), §09.19 (convergence summary).
These are the reasoning moves that made the progress believable and reproducible — shared at working-physicist depth so a reader can both check them and build on them.
(1) Two candidates, then computed elimination — "one dead, one survivor." Rather than positing an inflaton, the construction reads off the geometry's own moduli as the only candidate inflaton directions (ρ, σ), then eliminates by computation. The insight is that KK-stiffness is a target-blind discriminator: a modulus whose mass sits at the KK scale physically cannot slow-roll, independent of any cosmological comparator. This converts "which field inflates?" — normally a free choice — into a computed narrowing. The honest limit kept in view: the elimination is qualitative-grade, and a single survivor is narrowing, not a closed spectrum.
(2) Dial-free given the split — separating what the geometry fixes from what it does not. The decisive discipline is to isolate the one quantity that is dial-free given the frozen split (the plateau slope λ² = 4/K_σσ) from the quantities that carry hidden normalizations (the σ magnitude). The slope read-off has no adjustable parameter once the split is fixed — that is what makes the tilt direction a genuine geometric output. Crucially, the same discipline forces the honest fail-close: the forcing of K_σσ = 24 from the split is a physics claim that needs the frozen reduction source, so it is held as a theorem-candidate, not banked. The lesson generalizes: an arithmetic identity (4/24 = 1/6) is never itself physics; the physics is whether the inputs are forced.
(3) Turn a bookkeeping ambiguity into an observable. The normalization convention is exactly where a hidden tuning could hide. The insight that rescues honesty here is that the four named conventions produce r-bands that separate at the experiment's resolution — so instead of choosing a convention to land in the window (which would be true-by-construction and would RELOCATE the mystery), the construction pre-registers all four and lets the measurement read off which one nature uses. This is the cleanest possible disposition of an internal ambiguity: it becomes a falsifiable, convention-independent union prediction plus an information-bearing branch separation.
(4) Reduce-not-relabel, applied to the normalization. The most important anti-promotion insight in this gate: when the deep audit looked harder at K13-2, it did not find the single convention reducing — it found a second un-traced freedom (III-b, the (9/11)² factor) that the single-bit framing had hidden. The honest move was to demote: name the new freedom as a distinct posit and mark the band PROVISIONAL, rather than relabel the situation as resolved. Discharging III-b target-blind is strictly harder than the old framing claimed (it needs the external kinetic-metric reduction + the (9/11)² accounting + the 8/3 bridge, none reachable without fabrication), so the freedoms are NAMED, not relabeled away. This is the κ³/π discipline operating in real time.
(5) The κ³/π falsification test as a standing firewall. Every named convention/axiom the gate would rest on is checked against one question: would it be written WITHOUT knowing the comparators (Planck n_s, the r bound, N_eff, the LiteBIRD window)? The dial-free slope and the ρ-DEAD verdict pass cleanly (no comparator enters). A normalization branch chosen to land r in the LiteBIRD window would FAIL — it is true-by-construction and RELOCATES the mystery into the convention. This firewall is non-axiomatic (it is a verification procedure, not a measured invariant) and it guards III-a/III-b specifically.
(6) Scope honesty as a first-class result. The single highest-order insight: disclosing the excluded-sector wall up front is not a hedge — it is the binding frame. Because cosmology is out of scope (Axiom I / SG-10 §9.3.2), the gate's genuine wins are narrowing wins inside a diagnostic that can never close in-scope. Naming this prevents the most seductive overclaim ("the geometry predicts the CMB spectrum") and reframes the honest edge as a strength: a confident, falsifiable bet built to die under one measurement.
Insights distilled from 01_DOSSIER.md §2.2, §5; specialist_reply.txt §1, §4 (reduce-not-relabel, lines 139–142); ch_09_gap08_inflation.md §09.13(iii) ("the convention becomes an observable"); regrade …2026-06-29.md line 1413 (the overclaim corrections that motivated the scope-wall disclosure).
dcc66f1b2685 / manifest meta a5b1e6f9d951 — READ-ONLY; nothing in this gate mutates or closes it.| # | Witness | Grade | Holds? |
|---|---|---|---|
| W1 | Dimension split n = (6,2,1) fixing K_σσ | hand-checkable | Yes — facet of frozen branch |
| W2 | Plateau slope λ² = 4/K_σσ = 1/6 (dial-free given split) | symbolic | Yes as arithmetic; forcing fails-closed at A1-GEOM (CANDIDATE) |
| W3 | ρ-inflaton DEAD (KK-stiff) | symbolic | Computed verdict, qualitative grade; countersign pending |
| W4 | σ plateau → n_s ∈ [0.9643, 0.9679] | machine-lane | PROJECTED; conditional on K13-2 + countersigns + N_* |
| W5 | r ∈ [3.5, 10]×10⁻³ (operative branch T6) | machine-lane | PROJECTED, convention-anchored; moves to ~36×10⁻³ on other branches |
| W6 | K13-2: 81/22 = (11/2)·(9/11)² | hand-checkable | Arithmetic Yes; the (9/11)² factor is un-source-traced (III-b) |
| W7 | Four-branch table T6/T_u/D6/D_u | machine-lane | Asserted from §6.2 table; not independently re-run here |
| W8 | No-target-fitting firewall (κ³/π falsification test) | symbolic/audit | Conditional discipline claim; reviewer must confirm no comparator leaked |
| W9 | LiteBIRD union falsifier (r outside [3.5,36]×10⁻³) | symbolic | Yes — a genuine pre-registered falsifier (POSIT atomic; band provisional) |
| W10 | SG-10 excluded-sector lint | machine-lane | Yes in principle — boundary script-enforced |
Witness ledger reproduced from 01_DOSSIER.md §2.1.
ch_09_gap08_inflation.md §09.11).specialist_reply.txt §4 (lines 143–147).specialist_reply.txt §4 lines 148–150.)Reproducibility honesty: there is a small helper,
…/Fable_Version/rendered/scripts/inflation_sigma_map_read.py, associated with the σ-map yaml read (B3); the yaml file itself was not located/read in scope for this dossier, so the operative-branch bit is recorded as PENDING, not read. (Do not claim a read that did not happen.)
Binding ceiling for this whole section: no path closes the gate in-scope — cosmology is excluded (Axiom I / SG-10 §9.3.2). The realistic terminal bins are Reduced to Axiom (named convention), exported, BLOCKED, or OPEN (observation bet). Every numeric target below must pass the κ³/π falsification test (writable WITHOUT the comparators). A REFUTING result is a valid close.
(a) Precise statement. The physics claim "K_σσ = 24 is forced by n = (6, 2, 1), not chosen" is fail-closed. The arithmetic identity (λ² = 1/6 ⟺ K_σσ = 24) is trivially true; what is missing is the worked reduction showing K_σσ = 24 follows from the dimension counts of n = (6, 2, 1) under the frozen σ-kinetic metric, plus a stated relation between K_σσ = 24 (slope) and the K = 99/11 = 9 (normalization) — i.e. an account of the factor 24/9 = 8/3.
(b) Why it's hard / traps to avoid. The load-bearing line lives in the frozen σ-kinetic reduction source (the D.1 region, ≈ corpus line L10268), which is not in scope. The cardinal trap, named so it is not repeated: do not fabricate the reduction. A plausible-looking derivation that lands K_σσ = 24 would be the worst outcome (a fake "closure"). Second trap: do not silently assume K_σσ = 24 and K = 9 are the same object — the 8/3 factor must be accounted, not assumed away. Third trap: do not treat the small-integer coincidences (integer K = 9; the 8/3 factor) as evidence the gate banks — they are carried as open consistency questions / cited support only (the retired κ³/π and 5+3=8 cautionary discipline applies).
(c) Exactly what closes it. Read the frozen σ-kinetic reduction at L10268 in scope, target-blind, and (i) confirm or refute that K_σσ = 24 follows from n = (6, 2, 1) under the frozen kinetic metric, and (ii) state the K_σσ-vs-K relation (the 8/3 bridge). Success: a single value falls out target-blind ⇒ λ² = 1/6 becomes a banked theorem (or Reduce to Axiom AXIOM-BREATHING-KINETIC-NORMALIZATION, value-free, if the source merely records the convention). Refuting close (valid): the source does NOT single out K_σσ = 24 ⇒ λ² = 1/6 is a refuted candidate and the slope is a declared convention range. Either resolves the gate's central structural datum.
(d) Machinery & inputs. The dimensional-reduction kinetic-metric computation for a breathing modulus on n = (6, 2, 1); the frozen source at L10268 (Paper IV register). Start from ch_09_gap08_inflation.md §09.11 (the three-layer attempt + BLOCKER A1-GEOM statement) and the corresponding 03_attempt.md it references.
(e) Leverage. Closing this converts the σ slope from convention-anchored to geometry-forced — the difference between "convention-picked" and "geometry-forced" for the entire tilt direction. It is upstream of Holes B and C (the conventions) and the most leverage-bearing AI-pursuable theory move in the gate.
(a) Precise statement. The factor (9/11)² = 81/121 in 81/22 = (11/2)·(9/11)² is un-source-traced. It is mentioned by neither the breathing axis (B1) nor the normalization axis (B2), and a corpus-wide search for "81/121"/"(9/11)²" returns nothing in this context. The missing object is a source read establishing whether this factor follows from n = (6, 2, 1) target-blind or is a free convention.
(b) Why it's hard / traps to avoid. This is a newly-exposed freedom — the prior "one bit / one convention" framing hid it (the trap: believing the single-bit story and calling the band convention-independent). Until it is traced, the LiteBIRD operative band is PROVISIONAL-on-III-b, and the union's convention-independence holds only across the four named branches, NOT "regardless of any internal bookkeeping choice." Trap: do not restate the band as settled/byte-fixed while this freedom is open (this exact overclaim was caught in the regrade).
(c) Exactly what closes it. An external mechanical read of the frozen file at L10268 to source-trace 81/121. Success: the factor demotes to a derived consequence of the SHAPE-SPLIT (AXIOM-K13-2-RESIDUAL-NORMALIZATION-FACTOR discharged) ⇒ λ² = 1/6 becomes a banked theorem and the band becomes non-provisional. Refuting close (valid): the factor is a free convention ⇒ it stays a named value-free posit and the band stays a disclosed range. This is named the single highest-value AI-pursuable next move.
(d) Machinery & inputs. The frozen Paper IV register at L10268; the arithmetic anchor 81/22 = (11/2)·(9/11)² (re-verified exact). Start from specialist_reply.txt R1/§3 (III-b) and regrade …2026-06-29.md lines 1386–1387.
(e) Leverage. Resolving III-b is what makes λ² = 1/6 either a banked theorem or a refuted candidate — it co-closes with Hole A and removes the PROVISIONAL flag on the LiteBIRD band (Hole F).
(a) Precise statement. Which single band is operative (T6 vs T_u) is fixed by one mechanical read of the frozen σ-map yaml (the named K13-2 PENDING gate). The yaml + exact key + freeze hash are not located/read in scope.
(b) Why it's hard / traps to avoid. It is almost mechanical — but the trap is precisely to claim the read without the file in scope. "The gate is mechanical (no physics decision once the file is read)" — but only once read. If the frozen yaml is not produced, the item must fail-close, not be guessed. A second trap: the read must be non-circular — confirm the file is under a freeze hash so it cannot have been edited to match the bands it adjudicates (freeze-before-compare).
(c) Exactly what closes it. Identify the exact frozen yaml + key under a freeze hash, confirm it records K₆-only vs uniform breathing, and read the single bit. Success: the operative branch is fixed (decision-grade). Refuting/fail-close (valid): the frozen file is not produced ⇒ B3 stays PENDING and the operative-branch identity stays open. Confirm consistency with B1 (the D.1.0 uniform-breathing source declaration — TOE-GAP-08-Q06).
(d) Machinery & inputs. The helper …/Fable_Version/rendered/scripts/inflation_sigma_map_read.py; the frozen σ-map yaml (location TBD). Start from ch_09_gap08_inflation.md §09.3 and 01_DOSSIER.md §3.1 (R1/B3).
(e) Leverage. Fixes which of the four r-bands is operative (narrows the forecast from the union to a single band) and is an input to the absolute A_s (Hole D). Does NOT, by itself, close the gate.
(a) Precise statement. c_loop, the e^{−6σ} coefficient of V(σ), is OPEN. It gates only the absolute scalar amplitude A_s — NOT n_s, r, or the union falsifier. The missing object is the computed value of c_loop (sign + magnitude) inside the frozen window 0 < c_loop < A²/(4 c_bdry).
(b) Why it's hard / traps to avoid. It is a genuine 2-loop-equivalent FRG computation, AI-setup-able but NOT AI-closeable (the actual computation is the owner/machine FRG-2 lane). It is the Gap-04 residue — shared, so it should be exported to the moduli/loop specialist, not re-derived inside Gap-08. Trap: do not let c_loop be read as threatening the well — the v10 Amendment shows the well is unconditional in c_loop (c_bdry < 0 across its band), so c_loop's Gap-08 role is confined to the amplitude. Trap: confirm (don't assume) that n_s/r/union are c_loop-independent (TOE-GAP-08-Q09).
(c) Exactly what closes it. Run the pre-registered FRG-2 computation at the frozen radii (via the c_loop prereg runner) to return c_loop in the frozen window; then evaluate the frozen A_s relation and consistency-check against Planck A_s ≈ 2.1×10⁻⁹ (comparator-only, ±factor-2 fold tolerance). Success: c_loop in-window ⇒ A_s derivable. Refuting close (valid): c_loop outside the window, or an A_s mismatch beyond the factor-2 band ⇒ a recorded tension, not a silent re-tune.
(d) Machinery & inputs. FRG-2 method + operator basis (σ⁻⁶ vs the a₄/a₆ heat-kernel ladder of Gap-04 T2) + frozen-radii inputs (TOE-GAP-08-Q10); the c_loop prereg runner (v15, EXTERNAL). Start from C_c_loop_coefficient.md and ch_09_gap08_inflation.md §09.2.
(e) Leverage. Closing c_loop is the only block on the absolute amplitude A_s (κ₁₃ is already closed-by-computation), and it also supplies Gap-09's absolute scale. Cross-gate: it is the Gap-04 ↔ Gap-08 ↔ Gap-09 shared residue.
(a) Precise statement. N_ (the e-folds between pivot horizon-crossing and the end of inflation) is NOT STATED in the read corpus stanza, yet the n_s and r bands are evaluated at it. The bands' reproducibility is therefore incomplete. The missing object is the published N_ window + the explicit n_s(N_*, λ²) expression + the four-branch r-integration.
(b) Why it's hard / traps to avoid. N_ is set by the reheating history (Gap-09), so it cascades — it cannot be pinned inside Gap-08 alone. Trap: do not assert an N_ value to make the bands reproducible; publish the window and show consistency. Trap: do not present n_s as a prediction — it is a consistency-check (comparator held target-only).
(c) Exactly what closes it. Publish the N_ window the bands assume; show it is consistent with the Gap-09 reheating band T_RH ∈ [2.4×10¹², 4×10¹⁵] GeV; publish the explicit n_s(N_, λ²) and the four-branch r-integration so the bands are reproducible. Success: bands reproducible at a stated, Gap-09-consistent N_. Refuting close (valid): no N_ window is consistent with the Gap-09 band ⇒ a recorded cross-gate tension.
(d) Machinery & inputs. The slow-roll plateau map; the Gap-09 T_RH band and reheating→N_* matching. Start from ch_09_gap08_inflation.md §09.10 (A4), §09.12 (A2), §09.13 (A3); the Gap-09 chapter ch_10_gap09_reheating.md.
(e) Leverage. Completes the bands' reproducibility and ties the spectrum to the reheating history; couples to Hole D (A_s) and the Gap-09 re-gate (shared K13-2 + c_loop inputs).
(a) Precise statement. The σ-inflaton's life/death and the operative convention are observation-blocked: the LiteBIRD tensor-to-scalar measurement (~2030) has not been made.
(b) Why it's hard / traps to avoid. No computation substitutes for it — it is a contingent measured number describing the tensor amplitude our universe realized, the wrong shape of object for any axiom to dissolve. The model's job before ~2030 is to keep the four bands frozen (freeze-before-compare) so the test stays honest. Trap: do not re-tune the bands toward an emerging hint; a band moved to match data is true-by-construction and RELOCATES. Trap: remember the band edges are PROVISIONAL-on-III-b until Hole B lands — state the falsifier as atomic posit, provisional numeric band.
(c) Exactly what closes it (adjudicates it). The LiteBIRD measurement. r outside [3.5, 36]×10⁻³ ⇒ the candidate dies (a refuting result — a valid, honest close of the bet). r inside the union ⇒ the candidate survives and the measured value reads off which of the four named conventions is operative (the convention becomes an observable).
(d) Machinery & inputs. External experiment (LiteBIRD). The frozen four-branch table and union [3.5, 36]×10⁻³. Consistency question TOE-GAP-08-Q16: after the v15 sharpening of T_u to [6.1, 17], the union edges are unchanged (lower still 3.5 from T6, upper still 36 from D_u) — confirm.
(e) Leverage. This is the gate's headline adjudicator — but note: even a confirming LiteBIRD reading discharges the falsifier, NOT the scope-wall (Hole G). The gate cannot close in-scope whatever LiteBIRD returns.
(a) Precise statement. Cosmology (inflation / reheating / CMB / structure) is a declared out-of-scope sector (Axiom I / SG-10 §9.3.2). This is the gate's dominant terminal and the single highest-order reason it can never close in-scope.
(b) Why it's not a closeable hole. It is ATOMIC by kind — a definitional domain boundary, droppable only by enlarging scope, never derived. It is not a defect to fix; it is the binding frame. The trap (caught in the regrade) is to omit it — presenting the falsifiable spectrum as if a LiteBIRD confirmation would validate an in-scope cosmological claim. It would not.
(c) The correct disposition. Disclose it as the dominant terminal. Even a perfect n_s/r match remains a diagnostic consistency, never a derivation, and may never be cited to support an in-scope claim (SG-10 lint, script-enforced). There is no "close" here — only honest disclosure.
(d) Machinery & inputs. …/PER_GATE_DOSSIERS/SG10_COMPLETION_HANDOFF/01_DOSSIER.md §9.3.2; …/TOE/GAPS_AND_WALLS_REGISTER.md. The SG-10 lint enforces the boundary at the deploy step.
(e) Leverage. Disclosing it correctly protects the program's credibility: it converts the honest edge ("built to die") into a strength rather than letting a reader infer an in-scope cosmological claim that the program does not make.
These are not open weaknesses and must never be claimed as proven; they are universal negatives or unprovable-in-principle objects, correctly parked:
ch_09_gap08_inflation.md §09.7.)ch_09_gap08_inflation.md §09.6.)The dominant terminal. Gap-08 sits inside the EXCLUDED cosmology sector (Axiom I / SG-10 §9.3.2). The gate can never close in-scope; even a perfect n_s/r match — and even a confirming LiteBIRD reading — is a diagnostic consistency, never a derivation, and may never support an in-scope claim. Dissolved ≠ solved; selection ≠ derivation; given-E ≠ derivation-of-E; a captured log ≠ an independent countersign.
What is genuinely banked (target-blind wins). Exactly two: (1) the computed ρ-DEAD verdict (KK-stiff, qualitative grade — real candidate-field narrowing) and (2) the dial-free-given-the-split tilt DIRECTION λ² = 4/K_σσ = 1/6 at CANDIDATE grade. Plus the sub-win that κ₁₃² = 1/(π⁴ M_U⁹ M_Pl²) is closed-by-computation and the breathing axis is closed-by-source. Both target-blind wins are narrowing, not closure.
What is explicitly NOT claimed. - NOT that λ² = 1/6 is a derived/banked output — it is a THEOREM-CANDIDATE, fail-closed at BLOCKER A1-GEOM; the σ number does not reduce, it RELOCATES onto ≥ 2 un-traced conventions (III-a parity bit; III-b (9/11)² = 81/121) plus the unbanked 24/9 = 8/3 reconciliation. - NOT that n_s or r is a prediction in the reserved sense — they are PROJECTED outputs at CANDIDATE grade; the Planck n_s agreement is a consistency-check (comparator held target-only). - NOT a certificate-grade ρ-death — the qualitative KK-stiff verdict is the recorded terminal; the NT-1 χ-leg is VOID-both-directions (no value asserted). - NOT an absolute amplitude — A_s is BLOCKED on c_loop (the only block, κ₁₃ being closed). - NOT "convention-independent regardless of ANY bookkeeping choice" — convention-independence is banked only across the four NAMED branches; III-b lies outside them, so the numeric band is PROVISIONAL-on-III-b. - NOT that the small-integer coincidences (integer K = 9; the 8/3 factor; expressibility hit-counts) are evidence the gate banks — they are open consistency questions / unverified illustrative figures, carried as support, not proof.
The anchors paid. No measured invariant is an input: the comparators (Planck n_s, the r bound, N_eff, A_s) are held strictly as comparison targets (the ≥ 1 measured-invariant floor anchor; transfer-onto, not anchor-on). The convention axes are paid as named axioms: III-a at OWNER-DECISION grade (value-free, demotion clause live), III-b as REDUCE_FURTHER (un-source-traced). The κ³/π falsification test is a non-axiomatic discipline-firewall guarding those choices.
The honest ceiling, stated as the strength it is. Gap-08 is a serious candidate consistency — pre-registered to die. The geometry fixes a tilt direction and fails the rest closed; the spectrum is built to be falsified by a single ~2030 measurement, and inside the surviving band the conventions themselves become observable. That is a confident, falsifiable bet, openly bounded by an excluded-sector wall it can never cross. NOT a validated spectrum. NOT an in-scope result. STATUS-UPGRADES:0. Frozen branch dcc66f1b2685 / a5b1e6f9d951 READ-ONLY.
Dossier built by synthesizing and expanding the existing corpus: the brief GATE_BRIEF_GAP08.md; the completion-handoff dossier …/PER_GATE_DOSSIERS/GAP08_COMPLETION_HANDOFF/01_DOSSIER.md, 02_CURRENT_STATE.md, specialist_reply.txt; the self-close chapter …/TOE_Self_close_chapters/ch_09_gap08_inflation.md; the named-missing-objects + c_loop subsystem under …/gap_08_inflation_spectrum/; and the regrade/hole-queue entries in …/TOE/GATE_REGRADE_BESTCASE_AND_HOLES_2026-06-29.md. No number, band, hash, sign, or equation was invented; every quantity is traceable to a cited corpus file actually read, or marked OPEN. STATUS-UPGRADES:0.