Aesthetic Digital Twin

Nishi vs the field — every Nishi cell is measured against real organ source at emit time; each gap names the watch contract that will close it.

Nishi vs Infinigen 2.0 and MetaHuman 5.8 and Character Creator 5 and Anny and MakeHuman 1.3 -- not who renders the prettiest face, but whether the markers the primary literature says matter are EXPRESSED per body region from the bone out, MEASURED against a canon whose rows carry an evidence class, JUDGED by a referee that knows its fidelity tier, and provable for a NAMED LICENSED person. Measured 2026-08-27 across 37 axes: 24 are grounded in a named function that exists today and 13 are watch contracts. Zero rows claim an exceed. What we own is a parametric inside-out body whose 21 dimorphism coefficients are already data, a skull with named dimorphism knobs, seven face emitters, a silhouette ruler, a symmetry ruler and a one-lens referee; what we do not own is an evidence grade on any target, a curviness metric, a canthal tilt or lip ratio or cervicomental angle as measured axes, a rib-cage floor under the waist, a fidelity-aware referee, or a licensed real-person oracle.

Layer 1 · Executive

Where we are. Measured 2026-08-27, not asserted. WORKING: a parametric inside-out body whose 21 dimorphism coefficients are already data rows with a proven refusal; a generated skull with named brow, jaw, vault and zygomatic dimorphism knobs; seven face emitters (eye, nose, lips, chin, brow, ear, digits); a silhouette measurer with eye width as scale; one shared symmetry ruler; a skin albedo derived through the published ITA classification; an anatomy-refereed scalp groom; a volume-preserving muscle stage; tissue bound to an XPBD cage with a ringdown-identified plant; a real-person decision matrix; a referee panel with one voting lens admitted at fp 0 fn 0. OPEN, and this is the whole story: not one target row carries an evidence class, so a refuted number and a replicated one look identical to every consumer; there is no curviness metric, no canthal tilt, lip ratio, lower-third, cervicomental angle or cephalic index as a MEASURED axis, no rib-cage floor under the waist so an impossible torso emits in silence, no fidelity tag on the referee so a clay verdict and a photoreal verdict are the same verdict, and no licensed capture of any named person, so the operator's benchmark is today a refusal by construction. The brief's markers were checked against 41 pinned sources and split REPLICATED, CONTESTED, REFUTED; the board carries the grade, and the surgical ladder it arrived with is adopted as the region order and stripped of its clinic numbers.

Where we need to go. Every aesthetic claim this estate emits must be GRADED, MEASURED, JUDGED AT ITS FIDELITY and LAWFUL -- graded because each target row names its evidence class and its pinned source, measured because each body region has a numeric axis read from the emitted mesh rather than a slider position, judged at its fidelity because the referee refuses to compare across tiers, and lawful because a named person's marker vector is computed only behind a resolved licence chain and consent record. Infinigen, MetaHuman, Character Creator, Anny and MakeHuman appear as columns and bars only; Multi-HMR 2 and Anny appear as ORACLES only and never as product. The purpose the operator named is served in that order: prove the capability on graded evidence, then license likenesses for film, game, VR and print rather than copy them.

The unit. 1 u = one measured session-leg, the same calibration graphics.plan, procgen.plan, dcc.plan and anatomy.plan use, so costs are comparable across every board that touches a body. Fidelity claims are in millimetres through md_measure and permil through bb_detail_score and never through a new ruler; aesthetic claims are in NAMED AXES with a stated evidence class; a beauty number quoted without its fidelity tier and its evidence class is an opinion wearing a decimal point.

Where we already exceed on this board: 0 u. NOWHERE, and the zero is the measurement: beat_index 0 permil on the vs-ruler, a one-lens referee, and clay fidelity against published photoreal stills. The estate's real exceeds -- an inside-out generated body, integer determinism, a canon that is data, a referee that exists at all -- are priced on graphics, procgen, charsim and dcc and are not restated here.

Cost to a canon nobody can mistake for a slogan: 4 u. AT1 the evidence-graded target rows plus AT8 the fidelity-aware referee. Both are schema and calibration on incumbents, not new engines, and together they are the precondition for believing any number the later rungs produce.

Cost to a measured axis for every face and torso region: 9.5 u. AT2 curviness, AT3 averageness, AT4 canthal tilt and lip ratio, AT5 neoteny, AT6 cervicomental angle, AT7 the thoracic floor. This is the rung set that turns seven emitters and a silhouette into measured anatomy, region by region, in the order the brief's ladder ran.

Cost to a licensed real-person oracle that passes end to end: 7 u. AT10 the licence chain, AT9 the marker-vector oracle, AT14 the reverse-direction referee. The licence rung is first by law; the oracle and the referee are the capability the operator wants to prove, and the reason it can be sold instead of sued.

Cost to tissue-level fidelity and the male canon: 4.5 u. AT11 gynoid fat pooling (shared with procgen) and AT12 the male evidence rows. The first is the largest physical lift on the board and the one every hip, waist and breast claim ultimately rests on.

Cost to animals and fantasy through the same engine: 4 u. AT13, shared with procgen's species canon. The engine is species-agnostic by construction; the aesthetic canon for a non-human is an operator declaration of which human markers survive, and this board refuses to invent one.

Research bar. Infinigen 2.0.0a1 (Princeton, BSD-3) is measured on the procedural world and creature generator with five genome classes and interpolatable parts, and NO human generator by source read; releases/latest tag v2.0.0a1 read live 2026-08-27 and pinned in barcheck. Theirs: pinned bar, and the fork this board's brief names as the base [@infinigen23]. Ours: The bar for the creature and fantasy axis and the reason the estate's inside-out layered human is a differentiator rather than a copy: Infinigen creatures are single-surface..

Research bar. MetaHuman 5.8 (Epic Games) is measured on the photoreal far bar for face and hair, with guide-curve-driven grooms and a captured-performance batch API witnessed in the mirror. Theirs: UNPINNED by this mechanism because the repository is EULA-gated, measured on the procgen board [@metahuman58]. Ours: Best on periorbital, nasal, lip and hair sculpt fidelity in every column where it reads 2. Its slider space is artist-authored, which is the axis this board's canon differs on..

Research bar. Character Creator 5 (Reallusion) is measured on a closed commercial character creator with Smart Hair and a creator loop, witnessed through trade press with two struck claims left struck. Theirs: UNPINNED, closed product with no feed, NOT PROBED [@cc5-reallusion]. Ours: The creator-loop bar. Cited from trade press because the vendor page was never fetched, said plainly..

Research bar. Anny (NAVER LABS Europe, Apache 2.0) is measured on a scan-free parametric human calibrated against World Health Organization statistics with age, height, weight and muscle controls across all ages. Theirs: UNPINNED, repository NOT PROBED [@anny]. Ours: Best on population grounding and adiposity, and the ORACLE representation for the reverse-direction referee [@multihmr2]. Permissively licensed, which is why it can be an oracle at all..

Research bar. MakeHuman 1.3.0 (open source) is measured on the open target-driven parametric human whose base mesh Anny builds on; releases/latest tag v1.3.0 published 2024-05-15, read live 2026-08-27 and pinned in barcheck. Theirs: pinned bar [@makehuman]. Ours: The open reference for a slider space that can be inspected. Its targets are community-authored, so it is a bar for breadth and not for evidence..

28 of 41 capabilities measured|1 of them measured exceeds|13 open|coverage 682/1000|adoption 20 full / 8 partial

Layer 2 · Roadmap

Do this next — computed by the ranker, never chosen by a seat

Order from nx_compare_rank (nx_dr_ocm: (deficit + cost-of-delay + option + enables) x sponsor x self-sufficiency x momentum / cost). FINISH rows are rungs whose symbol is present but whose organ is short of full adoption: the cheapest closures on this board, listed before any new work. Stamp: # asof=1787883596 domain=aesthetictwin target_version=0.1 rungs=14 done=2 open=12 finish=0 ranker=nx_dr_ocm

#	Stage	Rung	Priority	Derivation
#1	0.1	Referee calibration across render fidelity (AT8) cjc_fidelity_calib	50	v=1 m=1 c=20
#2	later	Canthal tilt and lip ratio as measured axes (AT4) fa_canthal_tilt	1666	v=5 m=2 c=6
#3	later	Cervicomental angle of the neck (AT6) bp_cervicomental	1600	v=4 m=2 c=5
#4	later	Averageness against a population mean face (AT3) fa_averageness_axis	1333	v=4 m=2 c=6
#5	later	Male evidence rows in the canon (AT12) bp_shr_axis	800	v=2 m=2 c=5
#6	later	Outline curviness metric (AT2) am_outline_curvature	400	v=3 m=2 c=15
#7	later	Animals and fantasy through the same canon (AT13) bp_species_canon	307	v=2 m=2 c=13
#8	later	Licensed marker-vector oracle for a named model (AT9) rp_marker_oracle	266	v=2 m=2 c=15
#9	later	Neoteny and maturity blend axis (AT5) fa_neoteny_axis	200	v=1 m=1 c=5
#10	later	Person licence chain and media classes (AT10) al_person_chain	150	v=3 m=1 c=20
#11	later	Gynoid fat pooling and ptosis (AT11) sd_fat_pool	133	v=2 m=2 c=30
#12	later	Reverse-direction referee for the licensed oracle (AT14) rp_bench_report	100	v=1 m=1 c=10

Critical path — contract, done-rule, executor, cost

Rung	Closes with	Definition of done (pre-declared)	Executor	Est.
Evidence-graded aesthetic target rows (AT1) after	bp_evidence_grade	Every aesthetic target the canon carries becomes a row of knowledge/aesthetic_canon.conf with region, marker, band, evidence class from REPLICATED or CONTESTED or REFUTED, and the key of a source pinned in aesthetictwin.refs, consumed by nx_body_proc beside body_dimorphism.conf rather than by a second loader. Done-rule: a gate proves every row resolves its key against the refs register or is REFUSED by name, that a REFUTED row may be an axis to measure but is refused as a steering target, that the counts of the three classes SUM to the row population so the grade is a partition and not a headline, and a neg-control proves a row citing an unknown key does not load. The first rows written are the ones the matrix header already graded: adiposity REPLICATED [@tovee1999], averageness and female-face femininity REPLICATED [@lee2025], WHR 0.70 CONTESTED [@voracek2002] [@katzmarzyk2001] [@hubner2024], symmetry CONTESTED [@pentonvoak2001] [@vandongen2012], phi REFUTED [@naini2024], neoclassical canons REFUTED as prescriptions [@farkas1985].	Organ	2 u
Outline curviness metric (AT2) after AT1	am_outline_curvature	A curvature-based measure over the silhouette nx_anatomy_measure already extracts, reported beside WHR and never instead of it. Done-rule: a gate proves the metric ORDERS two silhouettes of equal WHR and different curviness the way the source's raters did [@hubner2024], that it is invariant to uniform scale, and a neg-control proves a straight-sided silhouette reads zero.	Organ	1.5 u
Averageness against a population mean face (AT3) after AT1	fa_averageness_axis	Derive the mean face from the canon's own dimorphism rows sampled over seeds, then a landmark distance from it. Done-rule: a gate proves the mean face scores zero, that the score is monotone in morph amplitude for each of the seven face morphs, and that a distinctiveness the source found to LOWER attractiveness [@lee2025] [@little2011] reads as a higher distance rather than being folded into a beauty number.	Organ	2 u
Canthal tilt and lip ratio as measured axes (AT4) after AT1	fa_canthal_tilt	Tilt in degrees from the medial and lateral canthus landmarks of the emitted mesh, banded by the source's classes [@lee2020canthal]; vermilion heights, the upper-to-lower ratio and the lip-area-to-lower-third ratio reported against the witnessed 1 to 2 and 9.6 percent figures [@popenko2017], the latter landing on fa_lip_ratio in the same organ. Done-rule: a gate proves both axes are read from mesh landmarks and not from slider positions, that a mirrored face reads the same tilt with opposite sign, and a neg-control proves a face with no eye landmarks reports UNMEASURED rather than zero.	Organ	2 u
Neoteny and maturity blend axis (AT5) after AT4	fa_neoteny_axis	An eye-area-to-face-area ratio, a lower-third height and a zygomatic projection combined into one axis on which a child-like and an adult-like morph land at opposite ends, carrying the measured sex differences in lower face height and palpebral fissure height [@babacan2025] and the trait definition the review literature uses [@lidborg2025] [@perrett1998]. Done-rule: a gate proves the two extreme morphs score at opposite ends, monotonicity along the blend, and that the axis is reported as a number with its inputs printed rather than as a verdict.	Organ	1.5 u
Cervicomental angle of the neck (AT6) after AT1	bp_cervicomental	The angle derived from hyoid and menton landmarks of the emitted body, monotone in the submental fat parameter, with the surgical anatomy of the axis carried as citation and no target [@ellenbogen1980] [@abdulshakoor2021]. Done-rule: a gate proves the angle falls as submental fat rises, that it is read from landmarks the canon already places, and a neg-control proves a body with no hyoid station reports UNMEASURED.	Organ	1.5 u
Rib-cage floor under the waist (AT7) after AT1	bp_thoracic_floor	The waist minimum DERIVED from the chest coefficients and the bone layer so a waist below the skeletal envelope is REFUSED by name and one at the floor emits. Done-rule: a gate plants a sub-floor waist and proves the refusal fires, proves a waist at the floor emits with the bone layer intact, and a neg-control proves the floor moves when the chest coefficients move rather than being a typed number.	Organ	1 u
Referee calibration across render fidelity (AT8) after	cjc_fidelity_calib	The panel carries a fidelity tag per image (silhouette, clay, textured, photoreal), calibrates floors per tier, and REFUSES a verdict that compares across tiers, because the optimum is known to flip with stimulus realism [@versluys2017]. Done-rule: a gate proves the same labelled corpus at two tiers yields two receipts with different floors, that a cross-tier comparison is refused by name, and that the existing single-tier receipt is reproduced bit-identically when every image carries the same tag so no current consumer changes.	Organ	2 u
Licensed marker-vector oracle for a named model (AT9) after AT1,AT2,AT4,AT10	rp_marker_oracle	Behind a resolved licence chain and consent record, compute a named model's marker vector (waist-to-hip, curviness, leg-to-body, canthal tilt, lip ratio, lower-third height, skin ITA) from licensed reference captures, publish a band per marker, and report per marker whether an emitted body lands in-band. Diora Baird is the named oracle class; the estate holds no licensed capture, so the oracle is a refusal today. Done-rule: a gate proves refusal by name without a licence row, proves under a synthetic licensed fixture that the vector is computed and the in-band verdict flips with the morph, and proves the vector carries the evidence class of each marker from AT1 so a CONTESTED marker cannot decide the benchmark alone.	Organ	3 u
Person licence chain and media classes (AT10) after	al_person_chain	BUILDS ON the phototwin lane's landed al_license_chain (chain row with owner, rights, term, territory, royalty and three named exits) rather than re-implementing it: the widening is the SUBJECT as owner of her likeness and the MEDIA CLASSES (film, game, VR room, print) as separately licensable rights rows, with the emit-for-publication path refusing without a resolved chain. Done-rule: a gate proves a publish path refuses by name without a chain, that a chain missing any conjunct names the missing one, and that the measurement path still runs so the oracle lane is not destroyed by the fence.	Organ	2 u
Gynoid fat pooling and ptosis (AT11) after AT7	sd_fat_pool	Fat that pools, sags and spreads on contact for the gynoid distribution -- waist, flank, hip, thigh and breast -- so the brief's contouring markets become a distribution parameter on a body that behaves. Shared with the procgen board. Done-rule: a gate proves volume conserved through pooling, that a higher fat parameter lowers the breast lower-pole point monotonically against the incompressible tissue truth [@breasttissue], and a neg-control proves a zero-fat body is bit-identical to today's emit.	Organ	3 u
Male evidence rows in the canon (AT12) after AT1	bp_shr_axis	The male body already emits at sex factor zero; this rung writes its graded rows: shoulder-to-hip ratio REPLICATED as more attractive when larger [@pazhoohi2023], leg-to-body about 0.5 standard deviations above baseline [@versluys2018], arm-to-body no effect [@versluys2018]. Done-rule: a gate proves a shoulder-to-hip axis is read from the acromion and trochanter stations, that the male rows carry classes and keys under AT1's schema, and that no female-row target leaks into the male emit.	Organ	1.5 u
Animals and fantasy through the same canon (AT13) after AT1,AT12	bp_species_canon	A second species and a stylised human through the identical canon, generator and emitter path, shared with the procgen board. The aesthetic canon for a non-human is an OPERATOR DECLARATION of which human markers survive; this rung refuses to invent one. Done-rule: a gate proves the species canon loads through the same loader, that a fantasy character's graded rows are a declared subset of the human rows rather than a new table, and a neg-control proves a species with no declared rows emits with no aesthetic verdict at all.	Organ	4 u
Reverse-direction referee for the licensed oracle (AT14) after AT9	rp_bench_report	A monocular mesh recovery used as an ORACLE ONLY: recover a metric body from one licensed photograph [@multihmr2] in the Anny representation [@anny], compute the same marker vector from the recovered mesh, and publish the disagreement between the two readings beside the benchmark, so a fidelity claim about a named person carries its reference uncertainty. Done-rule: a gate proves the recovered and emitted vectors are compared per marker, that the oracle-to-emit disagreement is a distinct published figure never folded into the score, and that the learned model is invoked only on the measurement path and is ABSENT from every emit path by the same fence as AT10.	Organ	2 u

Milestones

Milestone	Rungs	Cumulative
M1 · The canon is graded and the referee knows its fidelity	AT1,AT8	4 u
M2 · Every face and torso region has a measured axis	AT2,AT3,AT4,AT5,AT6,AT7	9.5 u
M3 · A licensed real-person oracle passes end to end	AT10,AT9,AT14	7 u
M4 · Tissue-level fidelity and the male canon	AT11,AT12	4.5 u
M5 · Animals and fantasy through the same engine	AT13	4 u

Layer 3 · Engineering

How this is scored. Every Nishi mark is measured: the generator reads the real organ source on disk and requires the implementing symbol to exist (no self-grading). A watching tag names the organ and symbol contracted to close a gap — the mark flips itself on the next compare beat when that workstream ships, and the comparewatch- plane row flips with it. The flip is necessary, not sufficient: it proves the symbol exists, never that the capability is good. The bar is the rung's pre-declared done-rule, proven by its gate — a symbol shipped without the behaviour behind it is a defect, and the flip is exactly what makes that defect visible instead of quiet. Competitor marks record documented capability presence — presence, not depth or scale. Adoption is measured too: every measured row carries where its organ stands on the estate's ladder (source → built → promoted → registered → invoked; libraries by importer reach minus validation importers; gates by the execution surfaces that run them). A row is fully adopted only at the top of its ladder; anything short is tagged partial with the exact remedy, so a build nobody promoted can no longer read as shipped. Census stamps: importers asof 1787849099, gate census asof 1787855507 (unix seconds; -1 = census absent).

Capability matrix — measured against source

◉ leads / measured exceed● present◐ partial○ absent · click any capability for its evidence

Capability	Nishi	Infinigen 2.0	MetaHuman 5.8	Character Creator 5	Anny	MakeHuman 1.3
target-canon-as-data-anthropometric-never-phi Measured: `bp_load_dimorph` exists in `runtime/_hdl_build/nx_body_proc.nx`, verified at emit. The canon is DATA: knowledge/body_dimorphism.conf carries the 21 coefficients (biacromial, biiliac, waist, chest, bust projection, neck, limb girth, muscle relief, stylisation) as per-mille of stature over four knobs, and the loader REFUSES a short file -- bite-proven 2026-08-14. What the rows do NOT carry is why each value is what it is. Anny is Best on population grounding because it calibrates against World Health Organization statistics rather than scans and says so [anny]; MakeHuman targets come from its anthropometric community [makehuman]; MetaHuman and Character Creator ship artist-authored slider spaces [metahuman58] [cc5-reallusion]; Infinigen ships no human generator at all, read from its own source and paper [infinigen23]. The two numbers this board refuses to hard-code are named here once: the golden ratio, for which there is no convincing evidence of a link to facial beauty [naini2024], and the neoclassical canons, which fit at most 40 percent of a normal population and must not be read as a prescription [farkas1985]. Adoption: `LIVE` — fully adopted (top of its ladder).	●	○	●	●	◉	●
evidence-grade-on-every-target-row Measured exceed: `bp_evidence_grade` in `runtime/_hdl_build/nx_body_proc.nx`, verified at emit. LANDED 2026-08-27, THE SAME DAY THE BOARD NAMED IT (rung AT1): knowledge/aesthetic_canon.conf carries 26 graded rows (7 REPLICATED, 17 CONTESTED, 2 REFUTED, 5 targets, partition summing) and bp_evidence_grade refuses BY NAME a row whose key does not resolve in this board's refs register and any REFUTED row offered as a steering target -- adiposity REPLICATED [tovee1999], averageness and female-face femininity REPLICATED [lee2025], WHR 0.70 CONTESTED [voracek2002] [katzmarzyk2001] [hubner2024] [lidborg2025], symmetry CONTESTED [pentonvoak2001] [vandongen2012], phi REFUTED [naini2024]. Proven by nx_aesthetic_canon_gate 9 of 9 with the verdict in the exit code, both refusals firing on runtime fixtures and both deny-guards carrying firing positive controls. EXCEED, earned in the declared order rather than asserted: the capability is live, nx_aesthetic_canon_gate reads 9 of 9 with the verdict in its exit code, and the mutation bite KILLED a compiling mutant (killed=1, not_reached=0) with the subject rebuilt and redeployed per mutant and both deployed twins restored byte-identical. No rival grades its sliders by evidence at all. Adoption: `LIVE` — fully adopted (top of its ladder).	◉	○	○	○	○	○
waist-hip-ratio-measured-never-a-0.70-target Measured: `nx_anatomy_measure_body` exists in `runtime/nx_anatomy_measure.nx`, verified at emit. The estate measures named axes from a silhouette with eye width as scale, so a WHR is readable today. It is an AXIS here, not a target, and the reasons are witnessed: the original claim rests on figure-drawing studies [singh1993] and an fMRI study of fourteen men viewing post-surgical bodies at about 0.7 [platek2010]; BMI was then shown to be the primary determinant with WHR overestimated through covariation [tovee1999]; centrefold WHR INCREASED with issue number over 577 issues (r 0.47) while BMI fell [voracek2002], and read 0.68 with no appreciable change over 240 centrefolds in a later window [katzmarzyk2001]; curviness predicts attractiveness better than WHR on matched drawings [hubner2024]; and a systematic review over 125,062 women found a LESS feminine WHR may cue past fertility and the evidence base too weak to support the reproductive-potential claim at all [lidborg2025]. The cross-cultural counter-evidence exists as a Nature letter on the Matsigenka [yu1998], cited here for existence only because its findings are not witnessed in our mirror. The estate's own ANSUR II parse of 2026-07-26 found zero of 1,986 women at or below 0.70 [ansur2]. Adoption: `LIB-GATE-ONLY importers=1` — PARTIAL: imported only by validation organs (gates, tests, benches): wire it into a shipping program. adoption LIB-GATE-ONLY importers=1	●	○	●	●	●	●
outline-curviness-metric Open — watching `runtime/nx_anatomy_measure.nx` : `am_outline_curvature`, re-measured on every compare beat. Ship that symbol and this mark flips itself; the `comparewatch-` plane row flips with it. The replacement the evidence names: a curvature-based measure of the body outline that predicts attractiveness better than the WHR it is usually taken to stand in for, on line drawings that vary curviness and width independently [hubner2024]. No rival exposes such a measure; every column is a category absence. Contract: a curvature integral over the silhouette the existing measurer already extracts, proven to order two silhouettes of EQUAL WHR and different curviness the way the paper's raters did. watching am_outline_curvature	○	○	○	○	○	○
adiposity-as-the-primary-body-axis Measured: `bg_parse_layers` exists in `runtime/_hdl_build/nx_body_gen.nx`, verified at emit. The one body cue the evidence puts FIRST is weight scaled for height, readable by an observer through the perimeter-area ratio of the outline [tovee1999]. Our fat layer is a real nested layer between muscle and skin so lean shows the abs and heavy hides them from the same muscle -- the exceed for that stack is priced on graphics and procgen, coded present here. Anny is Best on this axis: an explicit weight control calibrated on population statistics across all ages [anny]. Honest limit carried over: our fat only attenuates relief, it does not pool or sag, which is the fat-pooling watch below. Adoption: `REGISTERED-DARK` — PARTIAL: callable, authorised, no MCP invocation on record (a direct fork logs the runner, so this is not proof it never ran); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked); only referenced by nx_creatorfit -- a mention, not a run. adoption REGISTERED-DARK	●	○	●	●	◉	●
beauty-referee-panel-one-voting-lens Measured: `cjc_panel` exists in `runtime/nx_charjudge_census.nx`, verified at emit. The referee exists and its honesty is priced on graphics: a multi-lens panel where only the percept lens votes, calibrated on a 13-image union corpus at fp 0 fn 0, with symmetry and skin report-only and the VLM lens abstaining. No rival ships a referee at all, so the zeros describe a different product. What it cannot yet do is the row below, and the review literature it would have to encode lists the same predictors this board grades -- symmetry, sexually dimorphic shape, averageness, skin colour and texture [little2011]. Adoption: `LIVE` — fully adopted (top of its ladder).	●	○	○	○	○	○
referee-calibration-across-render-fidelity Open — watching `runtime/nx_charjudge_census.nx` : `cjc_fidelity_calib`, re-measured on every compare beat. Ship that symbol and this mark flips itself; the `comparewatch-` plane row flips with it. THE TRAP THIS ESTATE IS STANDING IN. On male figures the optimally attractive leg-to-body ratio was fractionally BELOW baseline for silhouettes and about 0.5 standard deviations ABOVE baseline for detailed computer-generated figures, with rater sensitivity lowest for silhouettes [versluys2017]. We render at clay fidelity, so a referee calibrated on photoreal stills and applied to our clay, or the reverse, teaches the wrong target. Contract: the panel carries a fidelity tag per image (silhouette, clay, textured, photoreal), floors are calibrated per tier, and a verdict that compares across tiers is REFUSED by name. watching cjc_fidelity_calib	○	○	○	○	○	○
bilateral-symmetry-ruler Measured: `go_image_symmetry` exists in `runtime/_hdl_build/nx_image_symmetry_lib.nx`, verified at emit. One shared symmetry core scores the gallery and the generate-N-keep-best ranker from a single source of truth. It stays REPORT-ONLY on the panel, and this board records why rather than hiding it: symmetric male faces were more attractive in older work [pentonvoak2001], symmetry was NOT a significant independent predictor once averageness and dimorphism were measured together [lee2025], and a meta-analysis over 44 effect sizes found the asymmetry-to-dimorphism link near zero with wide intervals [vandongen2012]. A lens the literature has not settled must not vote. Adoption: `LIB-WIRED importers=2 nonval=2` — fully adopted (top of its ladder).	●	○	○	○	○	○
averageness-against-a-population-mean-face Open — watching `runtime/nx_faceanat.nx` : `fa_averageness_axis`, re-measured on every compare beat. Ship that symbol and this mark flips itself; the `comparewatch-` plane row flips with it. The strongest replicated facial predictor is averageness, the converse of distinctiveness [lee2025] [little2011], and the estate has no mean face to measure distance from. MetaHuman, Character Creator and MakeHuman all ship a base mesh that is a de facto population mean [metahuman58] [cc5-reallusion] [makehuman], and Anny calibrates to population data [anny]; none exposes a distance-to-mean number. Contract: derive the mean face from the canon's own dimorphism rows sampled over seeds, then a landmark distance where the mean scores zero and the score is monotone in morph amplitude. watching fa_averageness_axis	○	○	●	●	●	●
femininity-dimorphism-axis-on-the-skull Measured: `sk_emit` exists in `runtime/_hdl_build/nx_skullgen.nx`, verified at emit. The dimorphism the evidence supports on FEMALE faces is expressed where it originates: the generated skull carries named brow, jaw, vault and zygomatic knobs each scaled by a sex factor and a robusticity factor, so a feminized shape is a parameter rather than a sculpt. The witnessed basis: subjects preferred feminized to average female face shapes across UK and Japanese populations, and the authors read the result as a selection pressure that limits sexual dimorphism and encourages neoteny [perrett1998]; femininity in female faces predicted attractiveness while masculinity in male faces did not [lee2025]. Rivals carry the same axis as artist morph targets on an authored mesh. Adoption: `REGISTERED-DARK` — PARTIAL: callable, authorised, no MCP invocation on record (a direct fork logs the runner, so this is not proof it never ran); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked). adoption REGISTERED-DARK	●	○	●	●	●	●
neoteny-and-maturity-blend Open — watching `runtime/nx_faceanat.nx` : `fa_neoteny_axis`, re-measured on every compare beat. Ship that symbol and this mark flips itself; the `comparewatch-` plane row flips with it. The brief's second marker, and the literature states the trait set without the slogan: a neotenous facial structure with large eyes, full lips and an oval face shape is what feminine morphology is typically taken to mean [lidborg2025], and preference for feminized shapes was read as encouraging neoteny [perrett1998]. No rival exposes a neoteny axis. Contract: an eye-area-to-face-area ratio, a lower-third height and a zygomatic projection combined into one measured axis on which a child-like and an adult-like morph land at opposite ends, with the sex differences in lower face height and palpebral fissure height carried as measured population facts [babacan2025]. watching fa_neoteny_axis	○	○	○	○	○	○
cranial-vault-cephalic-index Measured: `sk_write` exists in `runtime/_hdl_build/nx_skullgen.nx`, verified at emit. The vault half-breadth is a named skull constant with its own sex term, so the cephalic index (maximum width over maximum length times 100) is a derivable number today rather than a measured one. The brief called head shape Eurocentric; the primary literature says the CLASSIFICATION is population-specific -- a CT study of 520 Malaysian subadults had to define its own bands (dolichocephalic at or below 78.8, mesocephalic 78.9 to 89.0, brachycephalic 89.1 to 94.0) and found the index falls with age [hamdan2025]. The canon therefore carries the index as a measured axis with the population it was read from, never as a target. Adoption: `REGISTERED-DARK` — PARTIAL: callable, authorised, no MCP invocation on record (a direct fork logs the runner, so this is not proof it never ran); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked). adoption REGISTERED-DARK	●	○	●	●	●	●
periorbital-orbit-and-eye-emitter Measured: `emitEye` exists in `runtime/_hdl_build/nx_body_proc.nx`, verified at emit. An eye emitter exists and places a sphere at an anthropometric station; the estate also measures with eye width as its scale reference. MetaHuman is Best on the periorbital region by sculpt fidelity [metahuman58]. What the emitter does not carry is a canthal tilt, a crease height or a lid aperture as parameters, which is the watch below. Adoption: `LIVE` — fully adopted (top of its ladder).	●	○	◉	●	●	●
canthal-tilt-measured-axis Open — watching `runtime/nx_faceanat.nx` : `fa_canthal_tilt`, re-measured on every compare beat. Ship that symbol and this mark flips itself; the `comparewatch-` plane row flips with it. The brief asserted a target tilt of plus four to six degrees. The witnessed population fact is different in kind: in 76 Korean patients the mean canthal tilt read 8.45 degrees right and 8.42 left, class I about 3.2 degrees and class II about 9.6, with no class III present [lee2020canthal]. Contract: tilt measured from the emitted mesh's medial and lateral canthus landmarks in degrees, banded by the source's classes, never steered to a clinic number. watching fa_canthal_tilt	○	○	●	●	○	○
nasal-vault-and-nasolabial-angle Measured: `emitNose` exists in `runtime/_hdl_build/nx_body_proc.nx`, verified at emit. A nose emitter exists with a sex and stylisation input; it is crude against a MetaHuman-class sculpt [metahuman58] and says so on the graphics board. The angle the brief quoted as fixed is CONTESTED between two studies this board pins: surgeon raters put the ideal female nasolabial angle at 95.5 to 100.1 degrees [armijo2012], while a general-population sample put it at 100.9 to 108.9 [sinno2014]. The canon carries both bands with the population that produced each, and the alar width sex difference is a measured fact from a 100-subject morphometric study [babacan2025]. Adoption: `LIVE` — fully adopted (top of its ladder).	●	○	◉	●	●	●
lips-vermilion-and-upper-lower-ratio Measured: `emitLips` exists in `runtime/_hdl_build/nx_body_proc.nx`, verified at emit. A lip emitter exists with a fullness input. The brief prescribed a golden-ratio lip; the witnessed study that actually ranked lips found an upper-to-lower ratio of 1 to 2 most attractive and 2 to 1 least, with the most attractive surface area a 53.5 percent increase over baseline and equal to 9.6 percent of the lower facial third [popenko2017]. Phi is not in the result [naini2024]. Adoption: `LIVE` — fully adopted (top of its ladder).	●	○	◉	●	●	●
lip-ratio-measured-axis Open — watching `runtime/nx_faceanat.nx` : `fa_lip_ratio`, re-measured on every compare beat. Ship that symbol and this mark flips itself; the `comparewatch-` plane row flips with it. The measured counterpart of the row above: vermilion heights read from the emitted mesh, the upper-to-lower ratio and the lip-area-to-lower-third ratio reported as numbers against the 1 to 2 and 9.6 percent figures [popenko2017]. No rival reports either. watching fa_lip_ratio	○	○	○	○	○	○
mandible-width-and-chin-projection Measured: `emitChin` exists in `runtime/_hdl_build/nx_body_proc.nx`, verified at emit. Chin projection is emitted from a sex input and the mandible breadth is a named skull constant with its own sex term, so a narrow tapered jaw and a small chin are parameters. The witnessed basis for the sex difference is a lower face height that differs significantly between men and women in a 100-subject study [babacan2025]; the neoclassical vertical-thirds canon it is often measured against fits the fewest cases of any canon tested [farkas1985]. Adoption: `LIVE` — fully adopted (top of its ladder).	●	○	◉	●	●	●
ear-protrusion Measured: `emitEar` exists in `runtime/_hdl_build/nx_body_proc.nx`, verified at emit. An ear emitter exists. THIS ROW STATES A NAMED ABSENCE OF SCIENCE rather than borrowing a clinic number: the sweep behind this board found no primary attractiveness study of ear protrusion, and the otoplasty target the brief quoted came from a clinic page that does not meet the refs law. The canon carries ear angle as a measured axis with no target. Adoption: `LIVE` — fully adopted (top of its ladder).	●	○	●	●	●	●
cervicomental-angle-of-the-neck Open — watching `runtime/_hdl_build/nx_body_proc.nx` : `bp_cervicomental`, re-measured on every compare beat. Ship that symbol and this mark flips itself; the `comparewatch-` plane row flips with it. The neck coefficients exist (radius and its sex slimming) but no angle is emitted or measured. The witnessed literature is surgical: visual criteria for the youthful neck were set out in 1980 with anterior platysma adjustment and subplatysmal fat excision as the levers [ellenbogen1980], and a 2021 technical note describes patients under 40 asking for a more defined angle, treated at the floor of the mouth rather than the fat alone [abdulshakoor2021]. Neither is an attractiveness measurement, and the board says so. Contract: the angle derived from hyoid and menton landmarks, monotone in the submental fat parameter. watching bp_cervicomental	○	○	●	●	●	●
skin-tone-ita-from-the-genome Measured: `si_albedo_of_gene` exists in `runtime/nx_skin_ita.nx`, verified at emit. Skin albedo is derived per genome through the published Individual Typology Angle, computed in integer CIE Lab; the exceed is priced on graphics. The classification it implements is witnessed here from the primary review: ITA from L star and b star classifies skin into six groups from very light above 55 degrees through light, intermediate, tan and brown to dark, an objective and quantitative measure irrespective of ethnic origin [delbino2018]. The brief's skin-lightening market is therefore expressible as a measured ITA axis, and this board carries it as such with no target. Adoption: `LIB-WIRED importers=4 nonval=3` — fully adopted (top of its ladder).	●	○	●	●	●	●
scalp-hair-groom-and-texture Measured: `gg_stats` exists in `runtime/nx_nxa_groom_gate.nx`, verified at emit. A scalp groom exists with an anatomy referee (nose-arbitrated facing, density floor, length band, face-cone clearance) and renders as geometry on the game surface. MetaHuman is Best with joint-driven hair from guide curves [metahuman58]; Character Creator ships Smart Hair [cc5-reallusion]. The TEXTURE axis the brief's relaxer market targets -- curl radius, strand cross-section -- is absent from the groom entirely, named here rather than implied. Adoption: `GATE:LIVE trial=-` — fully adopted (top of its ladder).	●	○	◉	◉	○	●
ribcage-sets-the-waist-floor Measured: `bp_thoracic_floor` exists in `runtime/_hdl_build/nx_body_proc.nx`, verified at emit. LANDED 2026-08-27 (rung AT7): the waist can no longer be dialled below the model's own bone -- bp_thoracic_floor derives the minimum from the pubic ring the torso already emits (it moves with build through the dimorphism conf, proven by the gate's derivation tooth reading floor 36 to 43 when D13 doubles), the emit path refuses BY NAME with exit 6 below it, and the crotch emit call consumes the SAME function so the guard cannot drift from the geometry it guards. Declared imprecision, stated not hidden: the true lumbar floor is the vertebral column, which the canon does not yet carry; the pubic ring stands in until a lumbar row exists. Anny reads 1 because its population calibration bounds the slider space [anny]. Proven by nx_aesthetic_canon_gate 9 of 9 with a sub-floor fixture refusing and the real conf emitting. Adoption: `LIVE` — fully adopted (top of its ladder).	●	○	○	○	●	○
breast-shape-upper-to-lower-pole Measured: `emitF` exists in `runtime/_hdl_build/nx_body_proc.nx`, verified at emit. Bust projection is a dimorphism coefficient and the inframammary fold is a shape-basis row, so the region is expressed. The brief demanded maximal upper-pole fullness; the witnessed literature does not: 100 consecutive models chosen for the attractiveness of their breasts shared an upper-to-lower pole ratio of 45 to 55, a nipple angled upward at a mean of 20 degrees, a linear or slightly concave upper slope and a convex lower pole [mallucci2012], and a 2026 consensus of 80 surgeons plus 369 patients converged on moderate volume, pronounced projection, tight cleavage and a 50 to 50 pole ratio with no difference between European and South American surgeons [buccheri2026]. Two sources, two ratios, both carried. Adoption: `LIVE` — fully adopted (top of its ladder).	●	○	●	●	●	●
breast-tissue-mechanics-and-ptosis Measured: `dbs_displace_free` exists in `runtime/nx_dyna_bind_skin.nx`, verified at emit. Skin vertices are embedded in the XPBD cage and the plant is identified by ringdown inside published human bands; the exceed is priced on charsim. The subject truth it rests on is witnessed on that board and reused here: breast tissue modelled as incompressible with non-negligible anisotropy [breasttissue]. Ptosis as a function of age and volume is not modelled and is named in the fat-pooling watch. Adoption: `LIVE` — fully adopted (top of its ladder).	●	○	●	●	○	○
waist-flank-gynoid-fat-pooling Open — watching `runtime/nx_softdyn.nx` : `sd_fat_pool`, re-measured on every compare beat. Ship that symbol and this mark flips itself; the `comparewatch-` plane row flips with it. The same contract the procgen board watches, read here from the aesthetic side: the brief's Lipo 360 and hip-filler markets are a fat DISTRIBUTION parameter in a body that pools, sags and spreads on contact, which ours does not. Fat today only attenuates relief. One contract, two pages flip. watching sd_fat_pool	○	○	●	●	○	○
pelvis-and-gluteal-silhouette Measured: `emitR` exists in `runtime/_hdl_build/nx_body_proc.nx`, verified at emit. The biiliac half-breadth rises toward female by a named coefficient and the hip rings are emitted from it. The witnessed gluteal ideal from 1,146 respondents is a posterior WHR of 0.65 and a lateral WHR of 0.70 with the most prominent point at the vertical midpoint, and NO significant differences by age, gender or ethnicity of the respondent [wong2016] -- which is the evidence this board cites when the brief calls the target European. Adoption: `LIVE` — fully adopted (top of its ladder).	●	○	●	●	●	●
leg-to-body-ratio-band Measured: `bp_skel_load` exists in `runtime/_hdl_build/nx_body_proc.nx`, verified at emit. The trochanter station is a named constant of stature, so the leg-to-body ratio is a derivable axis. The evidence is REPLICATED and it is a band, not an extreme: mid-ranging female ratios were maximally attractive to 1,261 raters [frederick2010], an inverted U peaked at the average with higher values more attractive in females [kiire2016], and the male optimum sits about 0.5 standard deviations above baseline with arm-to-body ratio having no effect [versluys2018]. Adoption: `LIVE` — fully adopted (top of its ladder).	●	○	●	●	●	●
thigh-taper-and-calf-shape Measured: `emitLimbTube` exists in `runtime/_hdl_build/nx_body_proc.nx`, verified at emit. Limb tubes are emitted from the limb girth and muscularity coefficients and the gastrocnemius is a shape-basis row, so the region is expressed. NAMED ABSENCE OF SCIENCE: the sweep behind this board found no primary attractiveness study of thigh gap or calf shape, and the brief's calf-reduction and thigh-lipo targets are carried as measured axes with no target. Adoption: `LIVE` — fully adopted (top of its ladder).	●	○	●	●	●	●
foot-size-and-shape Measured: `emitDigitP` exists in `runtime/_hdl_build/nx_body_proc.nx`, verified at emit. Toes are emitted by rule. The one witnessed finding is cross-cultural and modest: small foot size was generally preferred for females and average size for males across nine cultures using line drawings [fessler2005]. The canon carries relative foot length as an axis banded on that result. Adoption: `LIVE` — fully adopted (top of its ladder).	●	○	●	●	●	●
named-muscles-under-the-skin Measured: `my_radius` exists in `runtime/_hdl_build/nx_myo.nx`, verified at emit. A volume-preserving muscle bound to bone exists as a pipeline stage with its own gate, so bulge and flex are computable; the anatomy census still reads the inner layers as coarse envelopes rather than named muscles, and the estate's own SOTA read says runtime FEM muscle is film-only and not the bar. Coded present because it is the layer every surface marker on this board sits on, and no rival ships a muscle layer under its sliders. Adoption: `REGISTERED-DARK` — PARTIAL: callable, authorised, no MCP invocation on record (a direct fork logs the runner, so this is not proof it never ran); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked). adoption REGISTERED-DARK	●	○	○	○	○	○
anatomical-oracle-mesh-ruler Measured: `bb_detail_score` exists in `runtime/_hdl_build/nx_bodybench.nx`, verified at emit. Silhouette IoU plus normal field plus a subdivision-proof detail judge against an oracle mesh; the exceed is priced on graphics and procgen. It is on this board because the words digital twin are only honest when a fidelity number exists, and this is the number -- with the anatomy board's caveat that it measures shape agreement, not anatomical identity. Adoption: `REGISTERED-DARK` — PARTIAL: callable, authorised, no MCP invocation on record (a direct fork logs the runner, so this is not proof it never ran); no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked); only referenced by nx_bodydensity_gate -- a mention, not a run. adoption REGISTERED-DARK	●	○	○	○	○	○
real-person-depiction-decision Measured: `rp_decision` exists in `runtime/nx_realperson.nx`, verified at emit. A decision matrix for depicting a real person -- likeness, minors, consent -- exists as code and is a policy page at every rival. It is the gate every row below stands behind. Adoption: `LIB-GATE-ONLY importers=1` — PARTIAL: imported only by validation organs (gates, tests, benches): wire it into a shipping program. adoption LIB-GATE-ONLY importers=1	●	○	○	○	○	○
licensed-marker-vector-oracle-for-a-named-model Open — watching `runtime/nx_realperson.nx` : `rp_marker_oracle`, re-measured on every compare beat. Ship that symbol and this mark flips itself; the `comparewatch-` plane row flips with it. THE OPERATOR'S BENCHMARK, made lawful: given a resolved licence chain and consent record plus licensed reference captures of a named model, compute her marker vector (waist-to-hip, curviness, leg-to-body, canthal tilt, lip ratio, lower-third height, skin ITA) and a band, then report per marker whether an emitted body lands in-band. Diora Baird is the named oracle class; the estate holds no licensed capture of her, so today the row is a refusal by construction. The reverse-direction referee is available as an ORACLE ONLY, never product: a monocular mesh recovery that returns a metric body from one licensed photograph [multihmr2] in the Anny representation [anny]. Done when the fixture proves refusal without a licence and an in-band verdict that flips with the morph under a synthetic licensed subject. THE SYNTHETIC FIXTURE IS NOW DECLARED AND MEASURED: the operator-chosen June Elara render [elararef], an adult fictional house generation whose GENREC is ledger-witnessed, reading percept 271 against the admitted referee's voting floor of 152 -- above the 228 bar graphics pre-declared for its own avatar done-rule -- with symmetry 488 and skin in the GOOD figure band, both report-only by this board's grades. It is the reference the marker vector is first computed FROM and the licensing pitch is demonstrated WITH. watching rp_marker_oracle	○	○	○	○	○	○
likeness-licence-chain-as-data Measured: `al_license_chain` exists in `runtime/nx_artlicense.nx`, verified at emit. LANDED on the phototwin lane 2026-08-24 (PT4) and coded PRESENT rather than left as a watch that would have flipped on a name: al_license_chain resolves a chain row (sku, owner, rights, start, end, territory, royalty permil, approved) with three named exits SHIP-OK, REVIEW and REFUSE, and a royalty ledger in integer cents beside it. The PERSON widening this board needs -- the subject as owner of her likeness and the MEDIA CLASSES (film, game, VR room, print) as separately licensable rights, with the emit path refusing without a chain -- is rung AT10 under its own watch symbol al_person_chain, so this cell cannot claim that work by carrying the incumbent's name. This is the mechanism that turns proving capability into licensing likenesses. Adoption: `REGISTERED-UNAUTHORISED` — PARTIAL: registered, no cap ever minted; no execution surface runs it either (clock, cron, daemon, roster, actlog and surfaced forks checked). adoption REGISTERED-UNAUTHORISED	●	○	○	○	○	○
male-body-through-the-same-dimorphism-knob Measured: `emitBrow` exists in `runtime/_hdl_build/nx_body_proc.nx`, verified at emit. Male is sex factor zero on the SAME 21 coefficients, so the male body already emits; what is absent is its evidence-graded rows. The witnessed male markers: larger shoulder-to-hip ratios were judged more attractive regardless of stimulus or rater sex, with neural responses modulated by the ratio for male but not female avatars [pazhoohi2023], and a leg-to-body ratio about 0.5 standard deviations above baseline [versluys2018]. Filed as the next canon tranche, not as a new engine. Adoption: `LIVE` — fully adopted (top of its ladder).	●	○	●	●	●	●
animals-and-fantasy-through-the-same-canon Measured: `creaturegen_build_plan` exists in `runtime/nx_creaturegen.nx`, verified at emit. Quadruped, bird and fish plans run through the same layered-anatomy method as the human, and the stylisation coefficients (anime head scale, leg scale, limb slimming) are already data rows. Infinigen is Best on creatures with five genome classes and interpolatable parts [infinigen23]. The aesthetic canon for a non-human is a design question this board names and does not answer: the evidence base above is human, and a fantasy character inherits it only where the operator declares which human markers it keeps. Adoption: `LIB-WIRED importers=9 nonval=3` — fully adopted (top of its ladder).	●	◉	○	○	○	○
male-evidence-rows-shoulder-to-hip-axis Open — watching `runtime/_hdl_build/nx_body_proc.nx` : `bp_shr_axis`, re-measured on every compare beat. Ship that symbol and this mark flips itself; the `comparewatch-` plane row flips with it. The measured male axis the canon owes: a shoulder-to-hip ratio read from the acromion and trochanter stations, carried under the evidence schema with its witnessed grade -- larger ratios judged more attractive regardless of stimulus or rater sex, with neural activity modulated by the ratio for male avatars only [pazhoohi2023]. The watch cell for rung AT12. watching bp_shr_axis	○	○	○	○	○	○
species-canon-through-the-shared-engine Open — watching `runtime/_hdl_build/nx_body_proc.nx` : `bp_species_canon`, re-measured on every compare beat. Ship that symbol and this mark flips itself; the `comparewatch-` plane row flips with it. One contract shared with the graphics board: a second species through the identical canon, generator and emitter path, which turns the species-agnostic claim into a measurement. Infinigen remains Best on creatures until it lands [infinigen23]. The watch cell for rung AT13. watching bp_species_canon	○	◉	○	○	○	○
reverse-referee-recovered-vector-vs-emitted Open — watching `runtime/nx_realperson.nx` : `rp_bench_report`, re-measured on every compare beat. Ship that symbol and this mark flips itself; the `comparewatch-` plane row flips with it. The licensed oracle's second witness: the marker vector recovered from one licensed photograph by the monocular oracle [multihmr2] compared per marker with the vector measured on the emitted body, and the oracle-to-emit disagreement published as its own figure so a per-person fidelity claim carries its reference uncertainty. The watch cell for rung AT14. watching rp_bench_report	○	○	○	○	○	○
person-licence-chain-and-media-classes Open — watching `runtime/nx_artlicense.nx` : `al_person_chain`, re-measured on every compare beat. Ship that symbol and this mark flips itself; the `comparewatch-` plane row flips with it. The PERSON widening of the landed chain resolver two rows up: the subject as owner of her likeness, and the MEDIA CLASSES (film, game, VR room, print) as separately licensable rights rows, with the emit-for-publication path refusing by name without a resolved chain. This is the door the operator's licensing purpose walks through, and the watch cell for rung AT10. watching al_person_chain	○	○	○	○	○	○

Debt register

Id	Sev	What it is	Unblock
craft-whr-equality-tooth-pins-a-contested-number	4	The shipped craft-world figure pins waist over hip at 0.7 EXACT as an equality tooth on the charsim board, described there as the published female mean. This board grades 0.70 CONTESTED: the estate's own ANSUR II parse found no fit woman at or below it [@ansur2], centrefold WHR rose over five decades [@voracek2002], and the fertility signal it stands for is too weak to support [@lidborg2025]. A contested number is pinned by an equality in a shipped world.	Re-derive the figure's waist and hip from the dimorphism coefficients under AT1 so the tooth pins agreement with the CANON, not with a literal; do NOT retune it to a different literal, which would be the same defect with a new number.
symmetry-ruler-report-only-by-evidence	2	The shared symmetry core scores every gallery image and feeds generate-N-keep-best ranking, while the newest evidence finds symmetry non-significant as an independent predictor [@lee2025] and the older evidence finds it significant [@pentonvoak2001]. It stays report-only on the panel by design.	Keep it report-only and say so on every surface that prints it; revisit only if a source on this board changes the grade.
diora-band-is-eyeball-evidence	3	The only prior figure the estate holds for the named oracle class is a 2026-04-24 memory band (bust to waist about 1.4 to 0.65, tall, hip flare) read by eye from gallery images. It is class-5 evidence and it has been used as a prompt target.	AT9 replaces it with a licensed measurement; until then no board surface may print it as a measured figure.
no-primary-science-for-ear-thigh-calf	2	Three regions the brief's ladder targets -- ear protrusion, thigh taper, calf shape -- returned no primary attractiveness study in the sweep behind this board. Their rows say so.	Carry them as measured axes with no target; a future sweep that finds a primary source adds a row under AT1 and the grade changes with it.

Risk register

Risk	Likelihood x impact	Mitigation
Publishing a beauty claim from a refuted or contested marker	certain x high	Every consumer of the canon today sees a number without a class. A 0.70 target, a golden-ratio lip or a fixed canthal tilt would read exactly like a replicated axis. AT1 is the mechanism and it is first for that reason.
A likeness computed or emitted without a licence	possible x high	The operator's purpose is to license likenesses. A benchmark taken from an unlicensed capture, or a twin emitted without a chain, converts the capability into the harm the

On these two registers. Rows are declared in the domain's plan file and carry the debt id, which is the join key back to the sovereign debt plane — that plane, not this page, is the authority on state. Reconciling them automatically (the regen reading the plane and refreshing these rows) is a named, owed rung; until it lands, treat an id here as a pointer to look up, not a status to trust.

Evidence for every claim. This is not an opinion matrix: each capability we mark present is grounded in real source and proven by a live-forked gate. See the autonomously-generated evidence page → (per-axis source lines + executed self-test verdicts) or the machine-readable evidence.json.

Honest verdict. The coverage above is capability presence measured against source — not depth, scale, or polish, where mature rivals may lead. Exceeds are claimed only where a mechanism backs them. Every open gap is a watch contract: it names the organ and symbol that closes it, and this page flips the cell itself when that workstream ships.

Person · product · place — not yet measured for this domain

Every compare carries this layer. Declare knowledge/compare/aesthetictwin.ppp (rows surface|nishi or c1..c4|label|url|connect naming OUR live surface and each rival's front door), run nx_ppp_probe domain aesthetictwin, and this section fills itself on the next beat: the same ruler on both sides — privacy and CX (third-party hosts, tracker classes, cookies, security headers), design and longevity (design hygiene, computed WCAG contrast, render-blocking resources, unsized media, script weight, theme and motion queries), findability (landmarks, skip link, on-site search, breadcrumb, headings, internal links).

References

Beyond a link list. Every reference below resolves twice — the publisher's copy and, where banked, the estate's own non-rottable library mirror with a content pin — and carries its evidence class plus the exact claim on this page it grounds. Keyed marks like [key] in the matrix notes jump here. A dash means honestly absent, never assumed.

[naini2024] Naini FB. The golden ratio -- dispelling the myth. Maxillofacial Plastic and Reconstructive Surgery 46, 2 (2024). CC BY. publisher · read in our library knowledge/fetched/cmp_aesthetictwin_naini2024.xml · pin h502065fad7eb689b7db03e06ef2b9d7f08dc78bfd812a1ca756adae47cddd4f0 · accessed 2026-08-27 · published-paperGrounds: target-canon-as-data-anthropometric-never-phi. Witnessed in the abstract of the mirrored full text: there is no convincing evidence that the golden ratio is linked to idealized human proportions or facial beauty, and no evidence to support its use in orthognathic or facial aesthetic surgical planning. This is why phi appears on the board only as REFUTED.
[farkas1985] Farkas LG, Hreczko TA, Kolar JC, Munro IR. Vertical and horizontal proportions of the face in young adult North American Caucasians: revision of neoclassical canons. Plastic and Reconstructive Surgery 75(3), 328-338 (1985). publisher · read in our library knowledge/fetched/cmp_aesthetictwin_farkas1985.json · pin h3304a75749ccb8ffb4ad39857708d7a716f4d3ac179e3709d1d8bc21e1b5c724 · accessed 2026-08-27 · published-paperGrounds: target-canon-as-data-anthropometric-never-phi. ABSTRACT-ONLY witness: nine neoclassical formulas tested in 153 young adults; the two most often valid were horizontal (interorbital width equals nose width in 40 percent, nose width equals a quarter of face width in 37 percent), the vertical profile canons fit worst, and the canons do not represent average facial proportions so their interpretation as a prescription must be tested.
[tovee1999] Tovee MJ, Maisey DS, Emery JL, Cornelissen PL. Visual cues to female physical attractiveness. Proceedings of the Royal Society B 266(1415), 211-218 (1999). publisher · read in our library knowledge/fetched/cmp_aesthetictwin_tovee1999.html · pin h394b2b93d9a2246c7323e1d40e6fe463783f47eda171cc22bc417640d60f5b54 · accessed 2026-08-27 · published-paperGrounds: adiposity-as-the-primary-body-axis. Witnessed in the mirrored article: weight scaled for height (BMI) is the primary determinant of sexual attractiveness rather than WHR, covariation of apparent BMI and WHR in earlier studies led to the overestimation of WHR, and the perimeter-area ratio of the outline is a reliable visual index of BMI. Free to read, not licensed for reuse; mirrored as evidence only.
[ansur2] Penn State OPEN Design Lab (host). ANSUR II, the 2012 Anthropometric Survey of US Army Personnel -- 93 measures on 4,082 men and 1,986 women, public release 2017. publisher · read in our library knowledge/fetched/cmp_aesthetictwin_ansur2.html · pin h33d87e139f20ffbd17d197f413557a6d8cc6ff8b694c54da8cd40845422bece0 · accessed 2026-08-27 · datasetGrounds: waist-hip-ratio-measured-never-a-0.70-target. Fetched anew through the sovereign fetcher on 2026-08-27 so this row carries its own provenance-journal entry; the charsim mirror of the same page predates the journal and reads UNPROVEN there, so reusing it would have raised the fleet ratchet. Grounds the dataset behind the estate's own 2026-07-26 parse that found zero of 1,986 women at or below WHR 0.70; that parse is the estate's own class-4 run and is cited as such, not as a published result.
[anny] Baradel F, Bregier R. Anny -- a free-to-use 3D human parametric model for all ages. NAVER LABS Europe research blog, 2025. Apache 2.0. publisher · read in our library knowledge/fetched/cmp_charsim_anny.html · pin hf34c997530b459bd78b9969702266bd4afc6038ca21502e12df98e2c88d799f9 · accessed 2026-08-25 · published-paperGrounds: target-canon-as-data-anthropometric-never-phi. REUSED from the charsim board: Anny uses no scans, calibrates against World Health Organization data, and exposes age, height, weight and muscle controls -- the population-grounding bar for the canon rows and the reason its column reads Best on adiposity.
[makehuman] MakeHuman Community. makehuman -- the open-source 3D human modeller, GitHub repository page (server-rendered README). Latest release v1.3.0 read live from api.github.com on 2026-08-27. publisher · read in our library knowledge/fetched/cmp_aesthetictwin_makehuman.html · pin hcbcc678cb83ff6e25356df550348596f3783df38280231e93c083e9ed5dd2747 · accessed 2026-08-27 · source-readGrounds: target-canon-as-data-anthropometric-never-phi. The fifth column and one of the two pinned bars. Cited for existence and self-description of an open, target-driven parametric human; the anthropometric grounding of its targets is the community's, and Anny's own paper states that it builds on this base mesh.
[katzmarzyk2001] Katzmarzyk PT, Davis C. Thinness and body shape of Playboy centerfolds from 1978 to 1998. International Journal of Obesity 25(4), 590-592 (2001). publisher · read in our library knowledge/fetched/cmp_aesthetictwin_katzmarzyk2001.json · pin he3271d6b82b86e40c3dc7a9bceb04b531dfc42eaf90b4ab1207fc6267f9f2be4 · accessed 2026-08-27 · published-paperGrounds: waist-hip-ratio-measured-never-a-0.70-target. ABSTRACT-ONLY witness: 240 centrefolds, twenty-year means BMI 18.1 and WHR 0.68 with no appreciable change in either over 1978 to 1998, and 70 percent underweight. The window where WHR held; read beside voracek2002, where it did not.
[voracek2002] Voracek M, Fisher ML. Shapely centrefolds? Temporal change in body measures: trend analysis. BMJ 325(7378), 1447-1448 (2002). publisher · read in our library knowledge/fetched/cmp_aesthetictwin_voracek2002.html · pin h9daeef42143f7823edadc5ea2933d9afc33c0fedb767316f00289c7f255a59f7 · accessed 2026-08-27 · published-paperGrounds: waist-hip-ratio-measured-never-a-0.70-target. Witnessed in the mirrored article body: 577 consecutive issues from December 1953 to December 2001; bust and hip decreased while waist increased; body mass index (r minus 0.46) and bust-to-hip ratio decreased while waist-to-hip ratio (r 0.47), waist-to-bust ratio and an androgyny index increased. THE DIRECT REFUTATION of the brief's claim that centrefold WHR is invariant across decades.
[hubner2024] Hubner R, Ufken ES. Curviness is a better predictor of a woman's body attractiveness than the waist-to-hip ratio. Scientific Reports 14, 23081 (2024). CC BY. publisher · read in our library knowledge/fetched/cmp_aesthetictwin_hubner2024.xml · pin h52b7e523d314b642163e40f733162fe0ac60ff2246bfdffdfa6e1853becb64cb · accessed 2026-08-27 · published-paperGrounds: outline-curviness-metric. Witnessed in the abstract of the mirrored full text: the WHR and curviness are not uniquely related; on line drawings differing in curviness and width, curviness is the better predictor even with a simple curvature-based measure, so the WHR is a poor measure of body curviness. The contract for the curviness watch row is this paper's measure.
[lidborg2025] Lidborg LH, Boothroyd LG. A systematic review of the association between women's morphological traits and fertility. Evolutionary Human Sciences 7, e42 (2025). CC BY. publisher · read in our library knowledge/fetched/cmp_aesthetictwin_lidborg2025.xml · pin hf0f4f4f50d8d0f1fdfc691f5cdcafe4f6dab4d75fe425c8469528e050b1dabcf · accessed 2026-08-27 · published-paperGrounds: waist-hip-ratio-measured-never-a-0.70-target. Witnessed in the abstract of the mirrored full text: 19 articles, 68 effect sizes, total N 125,062; a LESS feminine WHR may cue past fertility; no article measured facial femininity; the evidence base is too weak to support the claim that feminine morphological traits are associated with reproductive potential. Also grounds the neoteny row's trait definition (neotenous facial structure with large eyes, full lips, oval face).
[lee2025] Lee P, Li J, Rafiee Y, Jones BC, Shiramizu VKM. Further evidence that averageness and femininity, rather than symmetry and masculinity, predict facial attractiveness judgments. Scientific Reports 15, 5498 (2025). CC BY. publisher · read in our library knowledge/fetched/cmp_aesthetictwin_lee2025.xml · pin h357f9c19bae5eb392e19b2c29b0b0bbdfa77a7f6bc2e6533f97311aa00bfbe4a · accessed 2026-08-27 · published-paperGrounds: averageness-against-a-population-mean-face. Witnessed in the abstract of the mirrored full text: shape symmetry, averageness and sexual dimorphism measured from face images and their independent contributions assessed with linear mixed effects models; attractiveness was significantly predicted by averageness in male and female faces and femininity in female faces, but not by masculinity in male faces or by symmetry. The load-bearing grade for the face rows.
[pentonvoak2001] Penton-Voak IS, Jones BC, Little AC, Baker S, Tiddeman B, Burt DM, Perrett DI. Symmetry, sexual dimorphism in facial proportions and male facial attractiveness. Proceedings of the Royal Society B 268(1476), 1617-1623 (2001). publisher · read in our library knowledge/fetched/cmp_aesthetictwin_pentonvoak2001.html · pin hf3270febf95bc7987e003220b4b25942327025629dde809293d29b5a5b85af42 · accessed 2026-08-27 · published-paperGrounds: bilateral-symmetry-ruler. Witnessed in the mirrored article: symmetric male faces are more attractive but not reliably more masculine, and symmetric faces possess characteristics attractive independent of symmetry that remain undefined. The older half of the symmetry disagreement, carried beside lee2025.
[vandongen2012] Van Dongen S. Fluctuating asymmetry and masculinity/femininity in humans: a meta-analysis. Archives of Sexual Behavior 41(6), 1453-1460 (2012). publisher · read in our library knowledge/fetched/cmp_aesthetictwin_vandongen2012.json · pin h32770ac7a662bf9cb0c4d32e58958ecb19e7ac3c4243fe624b6571ce2fe0e433 · accessed 2026-08-27 · published-paperGrounds: bilateral-symmetry-ruler. ABSTRACT-ONLY witness: 44 effect sizes from 16 studies, mean correlation between developmental instability and sexual dimorphism 0.04 (SE 0.05), facial masculinity or femininity unrelated to instability (r 0.01), and wide confidence intervals that make general conclusions premature. The reason the brief's genetic-shield reading of symmetry is graded CONTESTED.
[perrett1998] Perrett DI, Lee KJ, Penton-Voak I, Rowland D, Yoshikawa S, Burt DM, Henzi SP, Castles DL, Akamatsu S. Effects of sexual dimorphism on facial attractiveness. Nature 394(6696), 884-887 (1998). publisher · read in our library knowledge/fetched/cmp_aesthetictwin_perrett1998.json · pin hdd47d9ea6fb49acba8793247a463be78b8ac1b010b6a38654cab0d6a6dd83814 · accessed 2026-08-27 · published-paperGrounds: femininity-dimorphism-axis-on-the-skull. ABSTRACT-ONLY witness: subjects preferred feminized to average female face shapes across UK and Japanese populations, more strongly within population which indicates attractiveness cues are learned; feminized male faces were also preferred; the authors read a selection pressure that limits sexual dimorphism and encourages neoteny.
[little2011] Little AC, Jones BC, DeBruine LM. Facial attractiveness: evolutionary based research. Philosophical Transactions of the Royal Society B 366(1571), 1638-1659 (2011). publisher · read in our library knowledge/fetched/cmp_aesthetictwin_little2011.html · pin h5a9b7b6a97b0b778ae968b2e809dac40327f4eb50298fc58f0d28064aa94f57b · accessed 2026-08-27 · published-paperGrounds: beauty-referee-panel-one-voting-lens. Witnessed in the mirrored review: the facial characteristics that influence attractiveness judgements are symmetry, sexually dimorphic shape cues, averageness, skin colour and texture and cues to personality, with individual differences from hormones, own attractiveness, visual experience and social learning. The predictor list a referee panel must encode, read from the review rather than from the brief.
[singh1993] Singh D. Adaptive significance of female physical attractiveness: role of waist-to-hip ratio. Journal of Personality and Social Psychology 65(2), 293-307 (1993). publisher · read in our library knowledge/fetched/cmp_aesthetictwin_singh1993.json · pin h7a18109987a057e91c78c0e7a7428adddcb2e331d1a8b2a501fea529392c5f89 · accessed 2026-08-27 · published-paperGrounds: waist-hip-ratio-measured-never-a-0.70-target. ABSTRACT-ONLY witness of the ORIGINAL claim: minor changes in the WHRs of Miss America winners and Playboy playmates over 30 to 60 years, and men across ages preferring figures with lower WHR. Cited so the claim's source is on the board beside the work that revised it.
[platek2010] Platek SM, Singh D. Optimal waist-to-hip ratios in women activate neural reward centers in men. PLoS ONE 5(2), e9042 (2010). CC BY. publisher · read in our library knowledge/fetched/cmp_aesthetictwin_platek2010.xml · pin h09981b5ef382ee7372e097255682a363eafeeb143c530a567eaa01932e5c7920 · accessed 2026-08-27 · published-paperGrounds: waist-hip-ratio-measured-never-a-0.70-target. Witnessed in the abstract of the mirrored full text: fMRI of men viewing bodies surgically altered to an approximately 0.7 WHR with redistributed fat and relatively unaffected BMI; postsurgical bodies activated bilateral orbitofrontal cortex and anterior cingulate, and attractiveness ratings regressed onto nucleus accumbens activation. The brief's reward-centre claim, with its actual design and scope.
[yu1998] Yu DW, Shepard GH. Is beauty in the eye of the beholder? Nature 396(6709), 321-322 (1998). Letter. publisher · read in our library knowledge/fetched/cmp_aesthetictwin_yu1998.json · pin h95b040e082e13f2880336864adba6049fd3c018272ad74cb1ef2091121e54749 · accessed 2026-08-27 · published-paperGrounds: waist-hip-ratio-measured-never-a-0.70-target. EXISTENCE-ONLY witness: the core record carries the title, the Nature citation, indexing terms for Peru and cultural characteristics, and NO abstract, so this row proves that the cross-cultural test exists and grounds nothing about its result.
[versluys2018] Versluys TMM, Foley RA, Skylark WJ. The influence of leg-to-body ratio, arm-to-body ratio and intra-limb ratio on male human attractiveness. Royal Society Open Science 5, 171790 (2018). CC BY. publisher · read in our library knowledge/fetched/cmp_aesthetictwin_versluys2018.xml · pin hda62aac4358fb075290ad83584b6ebe8e168b9f5d3548d726846ab8a99cb6d5e · accessed 2026-08-27 · published-paperGrounds: leg-to-body-ratio-band. Witnessed in the abstract of the mirrored full text: the optimally attractive male leg-to-body ratio is approximately 0.5 standard deviations above baseline, intra-limb ratio has a weak effect, and arm-to-body ratio has no effect. Male figures, stated as such.
[versluys2017] Versluys TMM, Skylark WJ. The effect of leg-to-body ratio on male attractiveness depends on the ecological validity of the figures. Royal Society Open Science 4, 170399 (2017). CC BY. publisher · read in our library knowledge/fetched/cmp_aesthetictwin_versluys2017.html · pin h9ff86c5c7994cc71b294f6bda0ebd0c90ca77b97363a8452a6d614576602aaa1 · accessed 2026-08-27 · published-paperGrounds: referee-calibration-across-render-fidelity. Witnessed in the mirrored article: the optimally attractive ratio for silhouettes was fractionally below the baseline while for detailed computer-generated images it was approximately 0.5 standard deviations above, sensitivity was lowest for silhouettes, and the effects of limb proportions depend on the ecological validity of the figures. THE FIDELITY-FLIP FINDING that makes a per-tier referee calibration a contract rather than a preference.
[kiire2016] Kiire S. Effect of leg-to-body ratio on body shape attractiveness. Archives of Sexual Behavior 45(4), 901-910 (2016). publisher · read in our library knowledge/fetched/cmp_aesthetictwin_kiire2016.json · pin h1e9e9ba49a4ead12ce853e119a944409549429f30e0eee247f6c9dbb4c08de20 · accessed 2026-08-27 · published-paperGrounds: leg-to-body-ratio-band. ABSTRACT-ONLY witness: three-dimensional stimuli rated by 40 women and 40 men in Japan; values closest to the average ratio rated most attractive, an inverted U peaked at the average, and higher ratios more attractive in females than males.
[frederick2010] Frederick DA, Hadji-Michael M, Furnham A, Swami V. The influence of leg-to-body ratio on judgments of female physical attractiveness: assessments of computer-generated images varying in LBR. Body Image 7(1), 51-55 (2010). publisher · read in our library knowledge/fetched/cmp_aesthetictwin_frederick2010.json · pin hf699d54e0794af03da17c29162ab7a236290a6129b200be005c39ad40eed854f · accessed 2026-08-27 · published-paperGrounds: leg-to-body-ratio-band. ABSTRACT-ONLY witness: eight levels of female leg-to-body ratio within the typical human range rated by a London community sample of 207 and two US university samples of 940 and 114; mid-ranging female ratios were perceived as maximally attractive.
[hamdan2025] Syed Mohd Hamdan SN, Rahmat RA, Munusamy SM, Ibrahim N. New population-specific cephalic index standards for Malaysian subadults: prevalence, growth patterns, and clinical implications from a CT imaging study. Clinics 81, 100855 (2025). CC BY. publisher · read in our library knowledge/fetched/cmp_aesthetictwin_hamdan2025cephalic.xml · pin h7f247f08e04e03f2fd17889ba5caf68ba9cfc10f393fe69feaa8dc239406b00a · accessed 2026-08-27 · published-paperGrounds: cranial-vault-cephalic-index. Witnessed in the abstract of the mirrored full text: CI equals maximum cephalic width over length times 100 from 520 CT images; population-specific bands dolichocephalic at or below 78.8, mesocephalic 78.9 to 89.0, brachycephalic 89.1 to 94.0, hyperbrachycephalic at or above 94.1; and a significant negative correlation with age. The evidence that the index is population-classified, not a universal target.
[lee2020canthal] Lee JY, Choung HW, Choung PH. Diagnostic analysis of vertical orbital dystopia and canthal tilt for surgical correction. Journal of the Korean Association of Oral and Maxillofacial Surgeons 46(6), 379-384 (2020). CC BY-NC. publisher · read in our library knowledge/fetched/cmp_aesthetictwin_lee2020canthal.xml · pin hb3886160535c283384a5259253473563739c695218eeba8c54496b3041593cd6 · accessed 2026-08-27 · published-paperGrounds: canthal-tilt-measured-axis. Witnessed in the abstract of the mirrored full text: 76 Korean patients, mean canthal tilt 8.45 degrees right and 8.42 left, class I about 3.2 degrees, class II about 9.6, no

generated by nx_swcompare_matrix (sovereign NishiLang organ) from knowledge/compare/aesthetictwin.matrix · every Nishi cell verified against organ source at emit time · watch cells re-measured on every compare beat · zero JS, zero trackers