performance-casting git:20260921.99a06beA

Immutable. This exact content never changes and is served at /api/v1/blob/ebfb80be493d9464.

---
name: performance-casting
description: Search and test screen-role candidates through role-specific face, relative-scale and performance proofs, then present a source-pinned shortlist for user selection. Use when casting suitability or performance range is unresolved; character-art retains stable appearance authority.
---

# Performance Casting

Consume [Character Art](../character-art/SKILL.md) as the stable appearance authority. This skill searches/tests realizations of that design; it does not redesign face/body/costume to win a proof. Return appearance changes to Character Art and alignment to Production Design.

ROLE SALIENCE > AESTHETIC OPTIMIZATION. Modern screen appeal should amplify the
role's defining features. If prettier conflicts with more recognizably this role,
preserve the role. Aesthetic optimization may refine defining discriminants but
must never reduce their salience. Read [role salience and references](references/role-salience.md)
when reconciling a search or compiling an archetypal role.

Choose a person who can carry the scene, not a portrait whose styling supplies the
missing role. Read actual character/story/scene/director sources, current visual
specification, stylization policy, user preferences, previous search and approval
status. Preserve their fingerprints. Candidate scene interpretations can guide a
new search without rewriting adopted story or references.

Project the loaded Character Package into a RoleArchetypeProfile with the desired audience impression, its visible
carriers, body relationships, key scene tasks and failure modes. Presence dimensions
are authored for the role: moral insistence, comic uncertainty, maternal calm,
quiet threat and intellectual force need different faces and states. No gender,
age, ethnicity, beauty level, large body, aggression or tragedy is a global default.
Appearance expresses a fictional interpretation, not a scientific personality
inference. Static evidence can suggest range; it cannot establish acting ability.

Before a new search, read [character excavation](references/character-excavation.md) as a package-completeness audit. Missing answers return to Character External Driver; the following questions project existing package meaning and never authorize casting to invent it.
Connect personality with identity, rank/occupation, social position, historical or
cultural environment and other people's treatment. Read authored watchability, the shared
root of strength and limitation, fate contrast, counter-stereotype, performance range
and Character Art visual choices. Missing embodiment returns to its author. All ten questions must change a casting decision; short
answers or a reason why a contrast is inapplicable are preferable to generic essays.
Keep source-supported facts, project interpretation and casting implications separate.
`IDENTITY_LOST` and `ERA_CONTEXT_LOST` must both PASS: remove the name and check whether
occupation/status and situated values remain recognizable. Missing, failed or stale
excavation blocks new media generation. Legacy profiles remain readable, not spend-ready.

Casting does not impose charm. It tests the package's authored watchability in candidates.
DO NOT CAST THE ENDING INTO THE FACE: betrayal need not look deceitful, sacrifice
saintly, or tragedy melancholy. Seek the capacity to travel the arc. Regal presence
means learned authority, bodily confidence, spatial rights and others' response;
never facial divination, a fixed nose, or an assumption that kindness erases rank.

Read [proof design and handoff](references/proofs.md) to choose a face-first,
body-first, performance-first or reduced route. Explain omitted stages. For every
stage, name the one principal question and concrete pass observations before
spending. Split conflicting questions instead of packing face, height, acting,
armor and heroism into one full-body generation. Search archetypes through genuinely
different visible structures; adjectives and seeds alone do not establish diversity.

Before another search, read [appeal, discriminants and complete proofs](references/visual-discriminants.md).
Choose the role's AudienceAppealMode: being liked and commanding respect are different
possible attractions. Neither youth, refinement nor intimidation is a universal default.
SEARCH tests contrasting authored structures and permits strong individuality or failure;
SELECTION assesses role-specific appeal, screen suitability and drift from actual evidence.
Use `VisualCastingPlan` and the CLI `compile`/`check-projection` modes to trace source
choices into submitted text. Do not replace the compiled prompt with a new hand summary.
The compiler preserves authored choices; it does not infer facial anatomy from psychology.
Story-dependent claims cannot be face criteria. Calibration outputs are diagnostic only.
For new candidate selection pass `--visual-plan`: its appeal finding and complete proof
scope supplement the existing selection gate. Key social identities require SOCIAL;
a COMPLETE candidate covers FACE, SCALE and SOCIAL, with performance when the role needs it.
Legacy profiles/reviews remain readable; they do not prove the new proof duties fulfilled.

Face search isolates face and baseline presence with equal chest-up framing/light.
A scale claim requires shared-plane anchors and visible contact with the same ground;
one unanchored portrait cannot prove exceptional height. A performance comparison
pins the selected identity and test conditions, changes a role-specific playable
state, and checks what stayed constant. Keep costume a controlled variable until
casting selection; costume approval is downstream.

For roles whose identity depends on social relations, schedule SOCIAL after relevant
body proof and before performance comparison. Show peers or differently ranked people
sharing real space: attention, spacing and who may decide reveal position. This can
mean sovereign distance, local professional authority, or ordinary group membership;
the subject need not be the center or highest power. Explain omission for other roles.

Use `contracts/performance_casting.py` and `performance_casting.stage_brief` for
provider-neutral, JSON-schema-exportable handoffs. The [local CLI](scripts/casting.py)
validates profiles, emits stage briefs, checks prior reviews/budget and presents the
user-selection gate. It never generates, creates entities or approves candidates.
Host/MCP/harness projects each brief into a currently verified provider route and
compares the executed request with the brief before reporting adherence.

Review actual media separately for attractiveness, distinctiveness, role/presence
fit, cultural fit, drift and performance hints. Use PASS / SHORTLIST / PARTIAL /
REJECT with visible reasons, not totals. Do not infer quality from contract validity.
Record `conditions_preserved` from the actual image: identical requests do not prove
identical light, framing or expression. This finding must PASS before any stage can
advance. A harder shadow or a frown cannot substitute for the requested presence.
Face SHORTLIST can advance within the declared count; scale and performance require
PASS on their essential observations. Failed prior proof cannot be skipped merely
to complete the pipeline. A reduced route requires a reason in the original profile,
not an after-failure waiver. Stop the search under the user's count/budget rules;
artistic dislike is not a technical retry.

For authorized generation, use the qualified Host visual execution route and durable
media completion. Read live capabilities and quotes; unknown balance remains unknown.
Keep owned task IDs, exact identity inputs, sourceRef and full-byte hashes. Import
successful outputs with media.import_media, resolve/read back, and retain candidates
as PENDING_USER_REVIEW without replacing formal Character references or adoption.
Present matched contact sheets and originals first. Stage readiness and Host PASS
are never user approval. Existing ProductionDesignFreeze remains the only downstream
production gate; report missing approval rather than manufacturing it.

For source reads use `work.get_work`, `script.get_script`, `scene.get_scene`,
`asset.get_asset`, `asset.search_assets`, `asset.list_assets`, `media.get_media`
and `media.list_media`. Retained physical outputs use `media.import_media` followed
by `media.resolve_media` and full-byte readback. Do not use source/adoption writes
to persist the local casting profile. The tool catalog remains unchanged.

For an explicit visual-medium route, read [route-aware casting](references/visual-routes.md). The route is separate from the face-first/body-first proof order; an omitted route sidecar preserves a frozen legacy brief without asserting its medium. New route-specific design requires an explicit Work route.

For an explicitly authorized first whole-character CG candidate, use
`scripts/full_body.py brief` then `reserve`. This separate
`CHARACTER_FULL_BODY_CASTING` purpose preserves FACE/SCALE semantics and does not
require an already approved CG face. Read the current approved Work: its explicit
visualRouteBinding and characterCastingAuthorizations must pin the same Work
revision, source bytes, profile/plan/style/spec and user directive. The scope is
exactly one output and stop-after-first-result. Design compilation alone never
grants this scope. Reserve the exact qualified Host request before submitting it
once through the existing official image template. An uncertain submit consumes
the opportunity: recover the same job, never automatically submit another.
Complete retained media with the existing media_delivery path, bound to the Work
as a candidate with userAdoption=PENDING. No Character adoption, second image,
legacy stage pass, scene, video or audio follows from this handoff.

## Route-owned expression

Select DESIGN_NEUTRAL, SUPPORTING_CASTING or HERO_CASTING independently of visual medium. HERO_CASTING is legal for live action and CG. Use the route-owned expression template and formal Work bundle; do not reuse actor-audition conditions for heroic CG. A new candidate stays unapproved. Read [expression route contract](../../docs/expression-route-contract.md) when using character or action expression profiles.

For HEROIC_CINEMATIC_CG full-body compilation, require the mode-specific visual intents and the final paragraph source map described in the expression route contract. Audit the exact provider prompt for real visual contrasts and conflicting positive directions before reserving a paid candidate. A recovered prior image is not an approved identity reference.

Keep generic amplification, opt-in archetype traits and character-instance intent separate. For an approved amplitude reference, follow the ownership and non-copying rules in the expression route contract; verify storage recovery before any authorized paid regression.

## Character Package ownership

Given Character Package, produce/evaluate candidate appearance. characterPackageRef and characterPackageVersion are explicit inputs, with checksum. Extract casting implications from the package; do not discover fundamental identity. Missing package for a dedicated role blocks new execution.

Use the unified read-only `CharacterRepository.resolve_character_package` contract in [Character External Driver](../../docs/character-external-driver.md). Pin characterPackageRef, characterPackageVersion and checksum. Never create or overwrite Character Core in this Skill. Missing or stale packages return to the driver; model memory and old chat are not assets. DESIGN_REVIEW access does not authorize casting or production.

For an external-package-only whole-body test, use `characters.casting.compile_package_casting`, `executable_package_casting` and `reserve_package_casting`. Author a PackageCastingProjection from the loaded package: each visual paragraph pins exact package pointers and source fingerprints. It is a task-specific visual interpretation, never a new Character Core. This entry accepts no legacy instance profile or Work expression bundle. Keep the existing qualified image projection, single-use reservation and durable media completion. A legacy-profile read during this package-only task blocks execution; record the Host file-access audit. Old full_body entry points remain for archival replay and their separately authorized scopes, never a fallback for this entry.

## Character Embodiment ownership

Given Character Art + Embodiment, generate/evaluate only within separately authorized scope. Do not discover character soul or infer missing bodily meaning from archetype. Casting evaluates embodiment without revising it.

Read [Embodiment contract](../../docs/character-embodiment.md). Use the unified read-only `characters.embodiment.embodiment_handoff` with exact ref/version/checksum and route. Never create or overwrite embodiment in downstream execution. Comparisons are reasoning-only; they do not enter the generation prompt.

## Visual Medium compilation

Read [Visual Medium Contract](../../docs/visual-medium-contract.md). New package projections carry `visualMediumIntent` with independent medium, treatment, realism and casting-mode fields. Host paragraphs provide source-pinned character facts, never their own positive CG/actor-photography declaration. Use the shared compiler and retain its emitted source map. Require `verify_medium_compilation` PASS on the exact final prompt before provider projection; a hand-written “Generic CG Grammar” label is not compilation proof. Legacy labels use the explicit adapter and conflict gate. Do not change provider prompt optimization or thinking to implement a medium.

## Medium-first fact handoff

Consume structured, source-pinned Character Art facts and casting intent; do not author a second appearance or medium authority.
Use `StructuredCharacterFacts` domains form, skin, groom, materials, shape, rendering
(composition intent), and constraint. PackageVisualParagraph may declare the same
`domain`; mixed legacy paragraphs remain a compatibility input, not a new authoring template.
Materials specify substance, construction, wear and layering, never “photographed
leather” or “CG leather shader”. Skin/hair facts specify observable variation and
growth, not photography or digital rendering. Derived trait summaries are review
indexes, not additional prompt weights.

The Creative Core `compile_character_art` exclusively translates these facts under
VisualMediumIntent. Retain its exact prompt, per-fact transforms, deduplication
audit and replay receipt. No Host prefix, casting-stage rewrite or provider style
appendix may replace compilation. DESIGN_NEUTRAL controls baseline casting assessment. LOOKDEV_NEUTRAL controls
neutral presentation independently; HERO_CASTING may use it without losing role weight. Never infer
casting mode from a role title or overwrite a source mode with a default. Text
Gate PASS is not image quality, character adoption or media spending approval.

## Render style and presentation handoff

Consume the Character Art-owned renderStylization and its pinned source. Never
infer visible CG from HEROIC or GROUNDED_STYLIZED, and never select style to match
a provider. Keep castingMode separate from presentationMode: HERO_CASTING plus
LOOKDEV_NEUTRAL is valid. Require both `mediumGate` and `renderStylizationGate`
and full receipt replay before projection. Legacy CG with unspecified style is
not evidence for visible CG readiness. PHOTOREAL_DIGITAL_HUMAN is valid CG, but
fails a VISIBLE_FILMIC_CG image review when macro form/material/groom hierarchy
does not read clearly. Retain failure evidence; no retry or adoption is implied.