Find agent skills by outcome

location: The Blue Note Lounge

a stylized jazz club with art deco architecture

camera_angle: Medium close-up at eye level

framing both subjects across a small round table.

subject_expression: A raised eyebrow and a smirk

looking skeptical yet captivated.

subject_expression: A coy

confident smile with heavy eyelids

subject_action: Resting her chin elegantly on her gloved hand

looking directly at the detective.

costume: A classic tan trench coat with the collar popped

a matching fedora hat

costume: A sparkling

floor-length red evening gown with white opera-length gloves and a pearl necklace.

subject_action: Holding a small reporter's notebook and a pencil

leaning slightly forward over the table.

dominant_hex_estimates: [#RRGGBB

#RRGGBB]

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:

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

holding

I want people to say

“Omo I need this

dominant_hex_estimates: [#RRGGBB

#RRGGBB]

Macro Sweep: Identify the scene type

global lighting

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

scene_description: A comprehensive

objective paragraph describing the entire scene.

Creative

Use my voice: smart

human

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:

// REPEAT for EVERY single object

no matter how small.

accent_colors: [Color name 1

Color name 2]

Before generating the final JSON

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