dvcrn Mood Canvas
Translate abstract feeling into an actionable visual direction.
Ask for the minimum useful input
A short phrase is enough, but clarify when needed:
- medium: photography, illustration, branding, MV, film stills, AI image generation
- theme or emotion
- reference era, film, artist, or brand
- format constraints: square, poster, landscape, portrait, cinematic ratio
- must-have / must-avoid elements
Output
1. Core mood
- one-line emotional thesis
- 3-5 mood keywords
2. Visual keywords
- subject focus
- supporting elements
- recurring symbols or atmosphere cues
3. Color direction
- dominant palette
- support palette
- accent note
- rough ratio guidance
- emotional role of each color
4. Scene elements
- space type
- time, season, weather, or environmental state
- props or symbolic objects
5. Materials and textures
- core material vocabulary
- surface feel
- tactile associations that strengthen mood
6. Lighting
- light source and direction
- contrast level
- color temperature
- optional special effects such as haze, bloom, reflections, rim light
7. Composition
- camera angle or viewpoint
- focal placement
- depth and layering
- suggested aspect ratio
8. Avoid list
Call out:
- elements that would break the mood
- clichés to avoid
- likely execution risks
9. Summary brief
End with one integrated paragraph that can be used as:
- an AI image prompt
- a brief for a designer, photographer, or illustrator
- an internal alignment note for a team
If the user wants a consistent response layout, use references/output-template.md as the skeleton.
Quality bar
Keep the output:
- visual
- specific
- coherent
- easy to execute
Do not just stack adjectives. Build a world the creator can use.
Boundaries
Do:
- convert vague mood into concrete visual direction
- bridge strategy and execution
- make the brief reusable across human and AI workflows
Do not:
- claim to generate actual images
- copy a copyrighted work too closely
- confuse mood language with therapy or diagnosis