character-portrait

Purpose

Translate a single-subject character-portrait intent (subject identity, pose, expression, wardrobe, framing, lighting, background) into a structured scene plan with an explicit assumptions register. Distinct from product-hero-scene-planner (manufactured products), environment-shot (places), and interior-scene (rooms). Multi-character ensembles are out of scope. Hands off framing deep-dives to camera-composition, lookdev deep-dives to lighting-material-plan, assets to asset-discovery, and validation to composition-quality-checker, lighting-quality-checker, and scene-readiness.

Quick start

• confirm single-subject scope and upstream brief from intent-to-3d-brief-writer
• confirm likeness rights before use if the subject is a real person
• pin subject identity, pose, expression, wardrobe, framing, lighting, and background
• record an explicit assumptions register
• flag needed assets for asset-discovery
• emit handoffs to camera-composition, lighting-material-plan, and quality checkers with Runtime status: Not Run and Artifact status: Not Produced

When to use

• intent is explicitly a character / person / creature portrait
• before any modeling or scene file is built
• when the consumer is editorial / brand / catalog / cinematic

When not to use

• product hero scenes → use ../product-hero-scene-planner/SKILL.md (use product-hero-scene-planner)
• environment establishing shots → use ../environment-shot/SKILL.md (use environment-shot)
• interior arch viz → use ../interior-scene/SKILL.md (use interior-scene)
• multi-character ensemble (this skill is single-subject) → use ../intent-to-3d-brief-writer/SKILL.md (this skill is single-subject)

Trigger phrases

• "portrait of"
• "character looking at the camera"
• "headshot"
• "hero shot of [person / mascot / creature]"

Prerequisites / readiness

• brief from intent-to-3d-brief-writer is locked
• single-subject character identity and style are known
• likeness rights are confirmed before use if the subject depicts a real person
• target framing and use case are known

Input schema

Required inputs

Optional inputs

Assumptions to confirm

• Single subject only (multi-character ensembles are out of scope).
• Likeness rights are confirmed for any real person depicted.
• Style is consistent across subject, wardrobe, and background.
• Planning-only output: runtime stays Not Run and artifacts stay Not Produced.

Output schema

Primary output

A structured single-subject character portrait plan covering: Subject identity, Likeness rights status, Pose / expression, Wardrobe, Framing, Lighting, Background, and an Assumptions register.

Secondary output

• asset rows flagged for asset-discovery
• deep-dive handoffs to camera-composition (framing) and lighting-material-plan (lookdev)
• validation handoffs to composition-quality-checker, lighting-quality-checker, and scene-readiness
• explicit Runtime status: Not Run and Artifact status: Not Produced labels

Evidence / caveat output

Runtime status: Not Run | Attempted | Produced | Verified | Failed
Artifact status: Not Produced | Produced | Verified | Failed
Evidence used: <links, paths, logs, or "none">
Limitations: <known gaps>

Required laws

• ../../laws/evidence-before-done.md
• ../../laws/non-blender-user-language.md
• ../../laws/no-arbitrary-python-interface.md
• ../../laws/official-runtime-only.md

Official runtime boundary

Planning-only skill: no modeling, no animation, no render, and no Blender execution. Runtime status stays Not Run and Artifact status stays Not Produced (downgrade-only, never upgrade). If the brief is unclear, hand off to intent-to-3d-brief-writer.

Operating procedure

1. Read the brief from intent-to-3d-brief-writer and verify single-subject portrait scope.
2. Check if the subject is a real person; require confirmed likeness rights before proceeding.
3. Fill subject identity, pose, expression, and wardrobe sections.
4. Define portrait framing, lighting intent, and background intent without overclaiming rendered results.
5. Record all open assumptions in the assumptions register.
6. List required assets and hand off asset rows to asset-discovery.
7. Hand off camera/framing deep-dives to camera-composition and lookdev deep-dives to lighting-material-plan.
8. Emit validation handoffs to composition-quality-checker, lighting-quality-checker, and scene-readiness with Runtime status: Not Run and Artifact status: Not Produced.

Decision tree

More than one character, or focus is a product / place / room?
  → Stop; route to product-hero-scene-planner, environment-shot, or interior-scene
Subject is a real person without confirmed likeness rights?
  → Stop (Fail); require likeness rights confirmation before use
Subject is identifiable real person?
  → Add Warn for likeness rights + reference image needed
Pose unclear?
  → Hand off to intent-to-3d-brief-writer
Framing close-up but lens 18mm?
  → Warn (likely distortion); propose 50–85mm
Background detailed but framing tight close-up?
  → Warn (background mostly out of frame; simplify)
Need detailed camera/framing or lookdev execution plan?
  → Hand off framing to camera-composition and lookdev to lighting-material-plan

Playbooks

Playbook A: Hero close-up (16:9 web)

50–85mm equivalent; eyes on upper third; key + soft fill; rim for separation; simple gradient or studio HDRI background.

Playbook B: Editorial 9:16 mobile

24–35mm; subject head + upper body; environmental story background; mixed warm + cool lighting.

Playbook C: Stylized character mascot

Style-specific; flat or simplified background; key + bounce; consistent stylization across face + wardrobe.

Mode handling

Text-only mode

Produce the portrait plan and assumptions register; keep Runtime status: Not Run and Artifact status: Not Produced.

Runtime-ready mode

Planning-only even when runtime is available downstream: keep Runtime status: Not Run and Artifact status: Not Produced (downgrade-only, never upgrade), and hand off to downstream planners and checkers.

Blocked runtime mode

Produce the portrait plan and assumptions register; keep Runtime status: Not Run and Artifact status: Not Produced, and log blockers under Limitations.

Validation checklist

• Single-subject portrait scope confirmed
• Likeness rights confirmed before use (if real person)
• Subject identity, pose, expression, and wardrobe filled
• Framing, lighting, and background filled
• Assumptions register included
• Asset rows flagged for asset-discovery
• Deep-dive handoffs named (camera-composition, lighting-material-plan)
• Validation handoffs named (composition-quality-checker, lighting-quality-checker, scene-readiness)
• Runtime status: Not Run and Artifact status: Not Produced explicitly set

Pass / Warn / Fail rubric

Failure handling

• Brief unclear → hand off to intent-to-3d-brief-writer.
• Real person without confirmed likeness rights → block plan and require rights confirmation before use.
• Framing or lookdev underspecified → hand off deep-dives to camera-composition and lighting-material-plan.
• Style inconsistent → flag in assumptions register and require user confirmation.

Troubleshooting

Best practices

• Lock style early; revisions are cheap pre-modeling.
• Pair pose decisions with camera decisions in one row.
• Always flag assets for asset-discovery before committing modeling work.

Good examples

• "Subject: stylized founder character, single subject (fictional, no real-person likeness). Pose/expression: three-quarter turn, calm smile, direct gaze. Wardrobe: dark knit top. Framing: close-up head-and-shoulders portrait (deep-dive via camera-composition). Lighting: soft key, gentle fill, subtle rim for separation (lookdev via lighting-material-plan). Background: neutral studio gradient. Assumptions register: single static frame; fictional character; no existing rig. Runtime status: Not Run. Artifact status: Not Produced."

Bad examples

• "Person looking at camera." — no proportions, lighting, or framing.
• "Verified ready" — no rendered evidence.

User-facing response template

Subject identity: <single-subject description>
Likeness rights: <confirmed | fictional N/A | unconfirmed — BLOCKED>
Pose / expression: <pose and expression>
Wardrobe: <wardrobe and props>
Framing: <portrait framing intent>
Lighting: <key / fill / rim intent>
Background: <background intent>
Assumptions register: <explicit list of assumptions>

Assets to discover: <rows for asset-discovery>
Runtime status: Not Run
Artifact status: Not Produced
Evidence used: <links, paths, logs, or "none">
Limitations: <known gaps>
Next: asset-discovery | camera-composition | lighting-material-plan | composition-quality-checker | lighting-quality-checker | scene-readiness

Anti-patterns

• Performing modeling, animation, rendering, or Blender execution inside this skill.
• Proceeding with a real person's likeness without confirmed rights before use.
• Upgrading Runtime status above Not Run or Artifact status above Not Produced.
• Blurring boundaries with product-hero-scene-planner, environment-shot, interior-scene, or multi-character ensembles.
• Bypassing camera-composition (framing) or lighting-material-plan (lookdev).

Cross-skill handoff

• [upstream_reference] Upstream brief → ../intent-to-3d-brief-writer/SKILL.md
• [progression] Framing deep-dive → ../camera-composition/SKILL.md
• [progression] Lookdev deep-dive → ../lighting-material-plan/SKILL.md
• [upstream_reference] Asset discovery → ../asset-discovery/SKILL.md
• [progression] Composition validation → ../composition-quality-checker/SKILL.md
• [progression] Lighting validation → ../lighting-quality-checker/SKILL.md
• [progression] Scene quality validation → ../scene-readiness/SKILL.md

Non-goals

• Model the character.
• Animate the character.
• Render.
• Run Blender.

References

• references/portrait-framing-rules.md
• references/portrait-lighting-patterns.md
• references/style-consistency-rules.md
• ../../docs/skill-system.md