CR_0028_sunda_clouded_leopard.md

creatures/CR_0028_sunda_clouded_leopard.md

CR_0028 — the Sunda clouded leopard, the hole in the forest's noise

CANON SUBORDINATION — this document is PROPOSAL-TIER: it serves canon and never outranks it.
The canon: the CVD · the T1 foundation docs · the T0 registries (registries/) · the spine
(docs/spine/CH_*.md) · the region pages (_source/02_Tier_2_Region_Pages/). Authority order: docs/DOC_MAP.md § 0.
Canon served (scanned from this document's own citations — widen it by hand where it is thin):
T0_Creature_Roster · T3_Creatures_Tameable · docs/spine/CH_05.md · _source/02_Tier_2_Region_Pages/sumatra_java.md
READ THAT CANON FIRST — open it and derive from it before you build anything from this document.
If this document disagrees with canon, CANON WINS and this document is the defect — fix the
document, never the canon. Nothing here is applied until it is ratified into canon.

STATUS: PROPOSAL-TIER design sheet, written to docs/proposals/bestiary/BESTIARY_ELEVATION_DOCTRINE.md.

It cites canon rows and writes none. It carries a CANON CONTRADICTION FINDING at § 9 and does not resolve it.

1. identity

closed-canopy forest either side of it

HEADING, see § 9

folklore column on this row is empty, no supernatural claim is staged on it, and the chapter's care

architecture rests on the begu ganjang, Naga Padoha and the krasue being the only

mythologically-stylized beings in it [CH_05.md:267]. · primary_axis A5 age and state, spent as sign

row as the one deciding case for the whole PREDATOR reach; combat_facing FALSE

[CREATURE_DEMAND_RULED_STATE_L11.json @ CH_05 CR_0028; DIRECTOR_RULING_L11.json RULING_8]

L82-120, L322-337 · docs/spine/CH_05.md L12, L94, L116, L265, L267 · sumatra_java.md § 3 L172-180,

L179-180, L234-241, L354-364, L380-391 ·

build/3d/schema/anatomy_reference/QUADRUPED.json § sub_plans.carnivoran_digitigrade ·

docs/proposals/bestiary/sheets/CR_0009_endemic_insect_cohort.md

THE DESIGN PROBLEM, STATED HONESTLY. This row's entire mesh demand rests on one cell. It is not fought, it

has no boss row, it has no drop anchor, and it is one of the three animals the region page names as what

the highland herds are a threat FROM [sumatra_java.md:239-241]. The doctrine's ordinary answer to a

never-faced row is the ordinary asset path — but a predator whose whole ecology is not being seen needs a

design step or the build ships a cat on a branch and calls it an ambush predator. This sheet is that step,

and its subject is the ENCOUNTER'S NEGATIVE: what the forest does when the animal is in it.

2. base_species

RE-DERIVED FROM THE ANATOMY REFERENCE TABLE. The doctrine's § 2.1 inline licence is expired. BUCKET

COVERAGE: build/3d/schema/anatomy_reference/QUADRUPED.json § sub_plans.carnivoran_digitigrade covers

this row through its declared "generic canid/felid" scope rather than by a census row id, against a

*Canis lupus familiaris* reference taxon, with the table's own standing caveat that FELID-SPECIFIC

GONIOMETRIC RANGES ARE UNMEASURED and canine ranges transfer as BOUNDS with the substitution declared

[ANATOMY_REFERENCE.md § "What this table does not know"]. Everything the table gives is a bound; the

arboreal specialisations below are outside every bucket and are stated as inline and labelled.

long-bodied against the bucket's cursorial reference body

20-23, taxon-variable; clavicle vestigial or absent with the forelimb in a muscular sling, so the

scapula translates on the ribcage; the bucket's deform chain and 34-bone minimum

bucket's range is 20-23 and taxon-variable, so the count is corrected to that range. The design read

does not change — a tail AS LONG AS the head and body together is a proportion claim, not a segment

count, and it survives at 23 segments with longer vertebrae.

foot toward the rear, which is what lets the animal descend a trunk head-first and hang from a bough by

the hind feet alone; and the upper canines, the longest of any living felid relative to skull length,

with the jaw articulation built to open around them. The table's carnivoran sub-plan is a cursorial

canid body and describes neither.

pronation-supination ±25; carpus 32-196; hip 50-162 with abduction-adduction ±30; stifle 42-162; hock

39-165, never hyperextending; intervertebral thoracolumbar per joint at sagittal ±8, lateral ±6, axial

±4, with the spine bending SAGITTALLY and not laterally. This last one corrects the first issue's

"lumbar sagittal arc of about 55 degrees", which was a whole-span figure quoted where the table's

convention is per joint.

the widest in the felid family and the design's single most important number; hind-foot rotation toward

180 degrees at the specialised ankle; and the tail as a declared chain with its own inertia, the

balance system rather than an ornament.

run is the ROTARY gallop LH-RH-RF-LF that the table names as the felid gallop with two airborne moments

per cycle. Inline: this animal is a climber first and a walker second, moving along the undersides of

boughs as readily as along their tops, and it uses the gallop least of any cat in the arc.

it — and hanging INVERTED, which is a silhouette no other creature in the chapter makes

3. lore_source

subspecies", and the region page names the clouded leopard among the five species for which none of its

six audited records carries any region-scoped attestation, refusing under cited-or-cut to assert one

[sumatra_java.md:382-391].

this animal is named as one of the three the herds give something to be a threat to [sumatra_java.md:239-241];

the porhalaan's own stake for a badly chosen day is the party waiting outside the shelter, exposed to

wild animals [SRC_00158, p. 170].

herd-move and the wrong-day exposure seat [sumatra_java.md:359-364; CH_05.md:94 — the typed site row,

corrected from an earlier :95 which is that row's env line rather than its terrain-and-threat values].

place in a food web, and this sheet builds from exactly that and declares everything else as ours.

4. elevation

ONE-LINE SILHOUETTE READ: a forest that has gone quiet, and a shape on the UNDERSIDE of a branch that

opens a mouth too big for it.

REGISTER: REAL_FAUNA (TRUE_TO_LIFE), doctrine § 3D, and the conversion is deliberate. This sheet's first

issue spent A4 HYPERTROPHY on the skull, "built out until the head reads long, narrow and wrong for a cat

of that mass." That was elevation against canon on a row canon holds mundane, and § 3D names A1-A4 as the

signal that a sheet is in the wrong register. The departure is WITHDRAWN. The real *Neofelis diardi*

already carries the longest canines relative to skull length of any living felid and a gape approaching

100 degrees — the most frightening fact available here is TRUE TO LIFE, and building past it was

gilding a thing that did not need it. Budget: A5 and A9 only.

THE REPLICA TEST DISCHARGES BY CONTRACT AND LADDER, per § 3D.

LEGIBILITY CONTRACT — silhouette-first discrimination against every confusable neighbour in CH_05, at

sixty pixels, black-filled:

ambush felids in one chapter: the tiger is a GROUND horizontal carrying its head below its shoulder

line; this cat is a CANOPY horizontal whose defining pose is INVERTED under a bough. Above-or-below the

branch is the discriminator, and neither animal can make the other's silhouette.

effectively none.

no tail, the cat hangs as one body line plus a tail line.

and the design's contract is that the player gets it once.

STATE LADDER — the states a photograph of the species-typical animal does not contain, and on this row the

ladder IS the design because evidence is the player's only relationship with it:

head-first — a directional sign no other animal in the forest leaves.

down a wet trunk carries the trunk on its forelimbs.

PRIMARY — A5 AGE AND STATE, spent as SIGN. A resident individual writes a readable record at a height

nothing else in the forest works at, and the player reads the animal entirely through what it has left.

MECHANISM: wear and deposition on level geometry over a residency, with a declared freshness ladder.

SUPPORTING — A9 THE MARK, SPENT INVERTED. The cloud blotches are large, soft-edged and dappled-canopy

shaped, so this animal's identifying feature is that it HAS no identifying feature — the register's

identity read taken in the negative. The sheet gives it no second mark, because a creature carrying three

marks carries none and this one is built to carry zero.

WHAT CARRIES THE PRESENCE INSTEAD OF AN AXIS: the forest's own behaviour, which belongs to the encounter

rather than to the body and is built in § 6 — the insect bed going quiet and the quiet TRAVELLING, a

macaque troop going vertical and holding, a browse line stopping, a herd bunching without a sound. It is

stated here so no lane mistakes its absence from the axis list for its absence from the design.

EMISSION: none. It reads by occlusion and by silhouette against a canopy light already in the scene.

5. articulation_delta

narrowed-peripheral-field cost is WITHDRAWN with the A4 spend. The skull is the real animal's, and the

slow whole-head turn it produced is kept only where the real anatomy already gives it: a short-limbed

cat inverted under a bough turns its head slowly because it is hanging, not because its temporal mass

was enlarged.

with its own ROM, not an animation cheat. It is inline AUTHORIAL_ANATOMY because no bucket in the table

describes an arboreal carnivoran ankle. The inverted hang is a legal pose in the solver, and any pose

outside the declared rotation is a solver failure rather than a design intent.

inertia. It is the balance system, and on the underside of a bough it is the part that moves when

nothing else does.

caught hair — with a declared freshness ladder, because the whole point is that the player can date them.

the real animal's.

6. fight_grammar

WHAT THIS ROW STAGES, STATED AFFIRMATIVELY. combat_permission is EMPTY, there is no boss row, there is

no drop anchor, and the L11 ruled state carries combat_facing FALSE. This creature is not fought at

Ch 5. What it IS, is a PREDATOR PRESENCE the player is genuinely threatened by, and the threat lands on

something the player is responsible for: the herd they are walking with, on the corridor the region page

attests, on a night the porhalaan may have called badly.

THE PART ECONOMY IS NONE, AND THE SHEET DECLARES IT RATHER THAN LEAVING THE BLOCK THIN. There is no body

to break because there is no fight, and evidence is not a part economy. Naming NONE here is a data fact

the schema can eventually check, not a design gap.

THE TELLS BELONG TO THE FOREST, NOT TO THE CAT, and that is the whole grammar.

tell shapebody originone answer
the insect bed stops, and the stopping travelsnot the cat — the ambient cohort answering itclose the herd up and move to the canopy edge rather than deeper in
a macaque troop goes vertical and holdsnot the cat — the troopit is at their height and moving; its heading is where they are NOT looking
fresh downward claw-scoring at head height on a trunkthe animal's own head-first descent on rotated hind feetthis is a lie-up trunk; move the escort route off it tonight
a bough with the wrong profile beneath itthe cat, inverted, hanging on its hind feet with the tail as a second linethere is no answer left; the take has happened or it has not
the gapethe real skull's own masseteric architecture at full openingthe one frame the animal is fully visible in, and the design's contract is that the player gets it once

THE CROSS-CREATURE LESSON, NAMED, BECAUSE THIS SHEET REUSES A SHAPE THE LANE HAS ALREADY LANDED. The

doctrine requires that a shape reused across creatures with a different answer declare which lesson it

teaches, and the landed CR_0009_endemic_insect_cohort.md sheet already spends INSECT SILENCE as a tell

at Ch 2: where the polo's work has been, "the cohort goes SILENT. Not absent, silent: the sound stops and

the player notices a hole in the mix before they notice anything visual." Two creatures, one shape, and

the discriminator is MOVEMENT:

lake — and it is there before the player arrives and after they leave. It has an EDGE the player walks

across, and crossing back out of it restores the sound.

direction of travel is the information. A player who stands still and listens to a static hole learns

where something HAPPENED; a player who stands still and listens to a travelling hole learns where

something IS GOING.

heading, on a player who has three chapters of the first reading behind them. That is a deliberate

escalation of one vocabulary rather than a collision, and the audio implementation must keep the two

distinguishable — a static hole and a travelling hole cannot share one state.

ESCALATION, WHICH IS NOT A PHASE SYSTEM BUT IS STILL NAMEABLE. What escalates along a route is EVIDENCE

DENSITY: scoring on one trunk, then three; hair at one fork, then a kill cached in a second; and the

insect bed's silences getting closer together. A player can say "we are in its range now" before any

system tells them, which is the escalation test passed on a creature with no phases.

RELOCATION AND TURF. It owns VERTICAL space rather than ground, and its range is a set of trunks rather

than a patch of floor. It does not defend and it does not contest — it relocates, and the sign that it has

is that the evidence ladder simply stops one day and starts again a valley over.

WHAT CHANGES IF A PERMISSION LANDS. The sheet says this plainly so a later lane does not have to redesign

from nothing: the fight is already written above. The gape is the opener, the inverted hang is the

approach, the slow hanging head-turn is the built-in punish window, and the counter class is positioning

rather than parry, because nothing on this body is a meetable contact surface. Note the § 3D ceiling: if

that permission arrives with a folklore cell or a boss row, the REAL_FAUNA register lapses and this sheet

re-derives under the full grammar.

7. care_read

manufacture one. ORDINARY_FAUNA-class rows carry no collective claim, and a lane must not re-apply

collective protection to game animals [doctrine part five; the rule is carried verbatim in CR_0024's own

cultural_care_register cell for this chapter].

photographic-wildlife template will not produce on its own — the plate exists to specify the two poses

the entire design depends on, and a documentary cat on a branch is exactly the failure mode.

KEPT and it is a real loss with a real cost, and the sheet refuses the softer staging in which the

player always arrives in time. BOLD, checked explicitly: SHOULD IT BE FACED? The row carries tier

PREDATOR and a corridor full of prey, and RULING_8 named this exact row as the deciding case for whether

predation reaches scope at all. REFUSED ON THE CELLS for this leg — no permission, no boss row, no drop

anchor, combat_facing FALSE — and the refusal is reversible by measurement rather than by argument:

the encounter above is built and waiting if a permission lands.

CR_0009, the other-animals-answer reads, the evidence ladder and its freshness dating, the escort stake

on the herd corridor, the descent-trunk scoring and the polished lie-up fork are INVENTED BY THIS SHEET

as the game's own. The Sundaic subspecies status, the PREDATOR tier and mid-level trophic role, and the

herd-corridor threat are CITED. The gape, the canine proportion, the rotating hind foot, the tail

proportion and the cloud-blotch camouflage are AUTHORIAL_ANATOMY where the table is silent; the

skeleton, joint bounds, caudal count and gait formalism are CARRIED FROM THE ANATOMY REFERENCE TABLE as

bounds with the felid substitution declared. The first issue's built-out skull and its

peripheral-field cost are WITHDRAWN with the A4 spend and are no longer claimed.

8. plate_brief

Written affirmatively. Every line names what IS in frame.

canopy, hind feet rotated and gripping, tail hanging as a second longer line, cloud blotches breaking the

body against dappled light, seen from below.

true proportion, the jaw at its widest.

the scoring it leaves visible in the bark beneath it.

bark, and a cached kill in a crotch — three reads at one scale, daylight.

them, the forest holding its breath — the plate that establishes the stake as a place.

9. boarded_debts

trade_8_tameable is EMPTY with a notes` cell reading "tameable_status unspecified in source [FLAG]".

THE SOURCE ASSIGNS IT BY HEADING WITH NO INLINE OVERRIDE — THE SAME BASIS ON WHICH CR_0024'S DANGEROUS

WAS TAKEN. In T3_Creatures_Tameable § 5 Ch 05 the clouded leopard is listed at L331 under the heading

"Mid-level and TAME —" at L327, immediately after the orangutan (L329) and the sun bear (L330), both of

which the roster DID fill as TAME from that same heading; and CR_0024's own DANGEROUS cell comes from

the "Apex and DANGEROUS —" heading at L322 in exactly the same way, since its L324 entry states no

status inline either. TWO FURTHER SUPPORTS. (1) THIS ROW ALREADY CARRIES THE HEADING'S OTHER HALF:

trophic_role reads "mid-level predator", which is the "Mid-level" of the same heading — the

extraction took one half of the line and dropped the other. (2) HEADINGS ARE DEFEASIBLE ONLY BY AN

INLINE OVERRIDE, AND THE DOC PROVES IT: CR_0025 the Sumatran rhinoceros sits under the same "Apex and

DANGEROUS —" heading at L325 and its entry says "OBSERVE tier" inline, and the roster correctly carries

OBSERVE rather than DANGEROUS. The clouded leopard's entry carries no such override. Target:

@ CR_0028 :: trade_8_tameable = TAME, with the [FLAG] cleared, evidence T3_Creatures_Tameable

L322-331. This is a registry write and this lane does not make it.

tier_classification ALONE, and RULING_8 records it as the single deciding case for PREDATOR's reach. If

the TAME cell were filled, this row would have entered on the same self-evident basis CR_0026 and CR_0027

did, and the swing would not have turned on one ruling.

PERMITTED. Three of the chapter's predators, two blanks and one explicit value.

threatened.

folklore_behavior_substrate, folklore_attribution_source, intelligence_tier, husbandry_breedable,

integrity_visible_aesthetic_shift and vril_substrate_response are all EMPTY.

design's two forest tells name two cohorts that HAVE NO ROW IN THIS REGION OR THIS CHAPTER:

chapter_first_encounter CH_02. There is no sumatra_java insect-cohort row anywhere in the roster,

so the bed this sheet reads the silence in does not exist as a creature row for CH_05.

angkor_khmer. Neither is a sumatra_java CH_05 row, so the troop that goes vertical has no row

either.

region-ambient FEEDs at the build pass as "the caldera-lake fish and waterfowl, the Kedu paddy fauna

on the Borobudur approach, and the highland birds" — no insect cohort and no macaque troop. The

tells are therefore reading two presences that neither the roster nor the page has staged.

that triggers a silence and the trigger radius have no shared key, and

creature_vocalization_sfx_id_ref_array is empty here and is the wrong column for a cohort's

silence in any case. CR_0009's own sheet boarded the same schema gap at Ch 2; this is its second

instance, now with a row gap underneath it.

region page already carries on its own Section-16 verify list [sumatra_java.md:388-391]. Every ethological

line in this sheet is labelled AUTHORIAL_ANATOMY because that acquisition has not landed.

Generated by harness/site/structure_site.py — the URL path is the repo path. review root