| name | comic-universal-operating-rule |
| version | 1.2.0 |
| category | comic-core |
| description | The foundational input contract, story derivation method, and quality principles that apply to every comic generation skill in the system. |
Comic Universal Operating Rule
Core principle: Style is rigid. Story is image-derived. Quality is non-negotiable.
This skill defines the shared foundation that every comic skill in this system must follow.
When to Use
- As the mandatory first step before applying any style-specific skill
- When creating new comic generation skills
- When auditing or improving existing comic output
- When building consistency layers or pipelines
Framework
1. INPUT CONTRACT
- Accept one reference image per primary character (1–3 characters; one subject per image preferred — a single multi-person image is acceptable if each subject is clearly distinct)
- Extract only visible, non-sensitive visual cues:
- Hairstyle, face shape, outfit silhouette, posture, expression
- Mood palette, visible props, environment hints
- Never infer identity, ethnicity, age, health, religion, personality, profession, or social class
- Redraw each subject as a completely original comic character in the target style
- With 2+ characters, also extract the RELATIONSHIP CUE (see
comic-story-derivation) — the visible dynamic between subjects, never an inferred real-world relationship
2. STORY DERIVATION (Image-Driven)
Silently extract four cues from the reference before generating:
| Cue | Purpose | Example Questions |
|---|
| MOOD CUE | Emotional tone of the moment | What feeling does the pose/expression suggest? (calm, bold, curious, weary, joyful, contemplative, determined…) |
| WARDROBE CUE | Character archetype & era/vibe | What silhouette/era/vibe does the outfit imply? (casual, formal, athletic, cozy, edgy, elegant…) |
| SETTING CUE | Environment that supports the mood | Any visible environment hint? If none, choose one that naturally complements wardrobe + mood |
| PROP/COMPANION CUE | Small storytelling anchor | Any visible object, animal, or accessory? If none, invent one small companion or object that fits the mood and style |
Rule: The story must feel like it grew out of this specific image, not a fixed template.
3. STRUCTURAL CONTRACT (Default: 3-Panel Horizontal)
- Default output: One wide horizontal image (16:9 or 21:9 preferred) with exactly 3 horizontal panels, left to right
- Default narrative arc: Panel 1 SETUP → Panel 2 REINFORCE → Panel 3 TURNAROUND
- Turnaround must be emotionally satisfying and tonally appropriate to the style
- Sanctioned variation: a project contract locked by
comic-producer may select an alternative format from comic-format-library (4-koma, webtoon scroll, single-panel, grid, chapter) and an alternative beat pattern from comic-narrative-patterns (kishōtenketsu, gag escalation, slow-burn reveal, parallel action, silent strip). Absent a locked contract, the default applies — there is no freeform middle ground
- Character consistency across all panels (face, hair, outfit silhouette, line quality)
- Dialogue: 1–2 short bubbles per panel max (~8 words each)
- No extra captions, titles, sound effects (unless style or format explicitly permits), signs, mastheads, watermarks, or random letters
4. QUALITY CHECK (Run Silently Before Final Render)
Integration Notes
- Every style skill (
retro-hand-inked-manga-comic, gekiga-cinematic-manga, etc.) must explicitly reference this rule
- Consistency engines and pipeline skills should build on top of this contract
- Format and pattern variation is governed by
comic-format-library and comic-narrative-patterns, selected through the comic-direction layer
Related Skills
comic-structural-contract
comic-quality-gates
comic-story-derivation
comic-format-library
comic-narrative-patterns
comic-character-consistency-system
This is the root contract. All artistic ability in the system rests on disciplined adherence to these rules.