SuperAPI AI Directory

visual-narrative

Design & creation

View repository

Updated 10-20 04:56

About this skill

Visual Narrative — Visual Narrative Methodology

Expert knowledge used by the dialogue-writer and comic-editor agents when integrating dialogue and visual elements.

Why Visual Narrative Matters

In comics, dialogue accounts for only about 30% of the storytelling. The remaining 70% is conveyed through art, composition, expressions, sound effects, and white space. The less dialogue, the stronger the comic.

Speech Bubble System

Speech Bubble Types

Type Shape Use Prompt Instruction
Normal Dialogue Rounded ellipse Regular conversation "speech bubble"
Thought Cloud shape Inner monologue "thought bubble, cloud shape"
Shout/Emphasis Explosion/Jagged Loud voice, anger "burst speech bubble, jagged"
Whisper Dashed ellipse Low voice "dashed speech bubble, whisper"
Electronic Voice Rectangular border Phone, broadcast, AI "rectangular speech box"
Narration Rectangular box (top) Commentary, time markers "narration box, caption"
Emotion Symbol Symbol only ?, !, ..., heart Symbols without bubbles

Speech Bubble Placement Rules

  1. Reading order: Top-to-bottom, left-to-right (for left-to-right languages)
  2. Tail direction: The bubble tail points toward the speaker's mouth
  3. No obstruction: Bubbles must not cover important visual elements
  4. Maximum 3: No more than 3 speech bubbles per panel (split panel if exceeded)
  5. Use blank space: Placing bubbles in the top third of the panel is most natural

Sound Effect Typography

Sound Effect Library

Category Examples Context
Impact WHAM! CRASH! BAM! THUD! Punches, collisions, explosions
Movement WHOOSH~ SWISH! ZOOM Running, flying, spinning
Emotion Thump thump, Shiver, Tingle Heartbeat, fear, excitement
Environment RUMBLE, SPLASH, RUSTLE Thunder, water, wind
Daily Life CLICK, SCREECH, DRIP, SNAP Doors, brakes, drops, cameras
Silence ... (dead silence) Tense quiet

Sound Effect Design Rules

  • Size = Volume: Larger text = louder sound, smaller text = quieter
  • Tilt = Speed: Tilted text = fast movement
  • Color = Nature: Red = destruction, Blue = cold, Yellow = electricity
  • Position = Source: Place near the source of the sound
  • Font: Use a different font from body text; handwritten feel recommended

Show-Don't-Tell Principle

Show Through Art, Don't Tell Through Dialogue

Bad Example (Tell) Good Example (Show)
"I'm so angry!" Clenched fist, gritted teeth in close-up
"This place is dangerous" Cracked walls, bloodstained signs, warning lights
"A lot of time has passed" Changing seasons, grown trees, rusted objects
"I started to like her" Eyes drifting toward her, flushed cheeks
"This is a precious object" Trembling hands carefully clutching it

Emotion Visual Codes

Emotion Visual Code
Anger Forehead veins, shaking frame, red background, gritted teeth
Sadness Rain/snow overlay, dark tones, slumped posture, tears
Joy Bright tones, flower petals/light rays, wide smile
Fear Wide eyes, darkness, long shadows, trembling hands
Surprise White pupils, halo effect, startled posture, sweat drops
Tension Narrow panels, consecutive close-ups, dark-light contrast

Page Transition Techniques

Scene Transition Methods

Technique Description Use Case
Cut Immediate scene change, no connection Time/location change
Dissolve Two scenes overlapping in one panel Flashback, daydream
Wipe Diagonal/horizontal scene split Simultaneous events
Match Cut Object in Scene A matches object in Scene B Metaphorical connection
Color Shift Color change indicates time/mood shift Past (sepia) -> Present (full color)

Text-Image Density Balance

Scene Type Text Ratio Image Ratio Panel Count
Action/Combat 10-20% 80-90% Many small panels
Daily Conversation 40-50% 50-60% Medium uniform panels
Emotional Scene 10-20% 80-90% Few large panels
Exposition/Flashback 30-40% 60-70% Narration boxes
Comedy 30-40% 60-70% Large panel for punchline