PPQS
WILD CORPUS · github_awesome

PQS 60 (B) - prompt from raw.githubusercontent.com

Source: raw.githubusercontent.com · Scraped 2026-05-04 · Scored 2026-05-04

Score

B
60 / 80
gemma4:latest · local · pqs-v2.0 · canonical
Clarity10 / 10
Specificity10 / 10
Context10 / 10
Constraints10 / 10
Output format10 / 10
Role definition8 / 10
Examples1 / 10
CoT structure1 / 10

The prompt

{
 "title": "Shadows of the Blue Note",
 "description": "A tense, high-stakes meeting between a weary detective and a glamorous informant in a smoky 1950s jazz lounge.",
 "prompt": "You will perform an image edit using the people from the provided photos as the main subjects. Preserve their core likeness. Transform Subject 1 (male) and Subject 2 (female) into characters from a classic 1950s film noir. Subject 1 is a rugged private investigator, and Subject 2 is an elegant femme fatale. They are seated at a secluded booth in a dimly lit, smoke-filled jazz club. The image must be ultra-photorealistic, utilizing cinematic lighting to create deep shadows and highlights. The scene should look like a frame from a high-budget blockbuster movie, shot on Arri Alexa, highly detailed, with a shallow depth of field focusing on their intense interaction.",
 "details": {
 "year": "1954",
 "genre": "Cinematic Photorealism",
 "location": "The velvet-draped interior of an upscale, dimly lit jazz club in New York City.",
 "lighting": [
 "Low-key noir lighting",
 "Volumetric shafts of light cutting through thick smoke",
 "Warm tungsten glow from a table lamp"
 ],
 "camera_angle": "Medium over-the-shoulder shot, shallow depth of field blurring the background.",
 "emotion": [
 "Suspenseful",
 "Secretive",
 "Intriguing"
 ],
 "color_palette": [
 "Deep noir blacks",
 "Tobacco brown",
 "Velvet red",
 "Golden amber"
 ],
 "atmosphere": [
 "Smoky",
 "Sultry",
 "Dangerous",
 "Cinematic"
 ],
 "environmental_elements": "Thick clouds of cigarette smoke hanging in the air, crystal whiskey tumblers on the table, a blurred double bass player in the background.",
 "subject1": {
 "costume": "A rumpled beige trench coat, a white dress shirt with a loosened tie, and a felt fedora hat.",
 "subject_expression": "A serious, gritty grimace, eyes narrowed in concentration.",
 "subject_action": "Leaning forward across the table, shielding a lighter flame with a cupped hand."
 },
 "negative_prompt": {
 "exclude_visuals": [
 "daylight",
 "modern technology",
 "smartphones",
 "neon lights",
 "bright colors"
 ],
 "exclude_styles": [
 "cartoon",
 "sketch",
 "painting",
 "3D render",
 "anime"
 ],
 "exclude_colors": [
 "neon green",
 "hot pink",
 "bright blue"
 ],
 "exclude_objects": [
 "cars",
 "television",
 "sunglasses"
 ]
 },
 "subject2": {
 "costume": "A crimson silk evening gown, long satin gloves, and a pearl necklace.",
 "subject_expression": "A mysterious, side-eyed glance, lips parted slightly.",
 "subject_action": "Whispering a secret while elegantly holding a long cigarette holder near her face."
 }
 }
}

This prompt was scraped from a public source. The score reflects the input as written, not the quality of any output it produced. The AI input quality problem is the gap between what people type and what the model can act on.