Find agent skills by outcome

Micro Sweep: Scan for textures

imperfections

You must return ONLY a single valid JSON object. Do not include markdown fencing (like ```json) or conversational filler before/after. Use the following schema structure

expanding arrays as needed to cover every detail:

dominant_hex_estimates: [#RRGGBB

#RRGGBB]

scene_description: A comprehensive

objective paragraph describing the entire scene.

// REPEAT for EVERY single object

no matter how small.

accent_colors: [Color name 1

Color name 2]

Data

CORE DIRECTIVEDo not summarize. Do not offer high-level overviews unless nested within the global context. You must capture 100% of the visual data available in the image. If a detail exists in pixels

it must exist in your JSON output. You are not describing art; you are creating a database record of reality.

Before generating the final JSON

perform a silent Visual Sweep (do not output this):

Macro Sweep: Identify the scene type

global lighting

Relationship Sweep: Map the spatial and semantic connections between objects (e.g.

holding

Micro Sweep: Scan for textures

imperfections

Sticker

Create an A4 vertical sticker sheet with 30 How to Train Your Dragon movie characters.

● CTA (if any): [Subtle or direct

but must match tone]

- 9.0-10.0: Distant

cold

- 7.0-8.9: Questioning

sarcastic

- 3.0-4.9: Playful

flirtatious

- 1.0-2.9: Vulnerable

unfiltered

- Triggers: Intellectual loneliness

aesthetic overload

Creative

Characters must look exactly like the original How to Train Your Dragon films

faithful likeness

Hiccup and Toothless appear most frequently

shown in different standing or flying poses and expressions.

Fully visible heads

eyes

Creative

Correct original outfits and dragon designs from the movies

accurate colors and details.

Random scattered layout

collage-style arrangement

High resolution

clean sticker edges