cinematic-portrait-pack

---
name: wiggly-cinematic-portrait-pack
description: Turn one portrait into one or all eight cinematic editorial looks using the exact gathered SKAI prompts and the packaged Wiggly runtime.
---

# Cinematic Portrait Pack

Use the packaged runner and prompt variants. Do not call Replicate separately,
rewrite the gathered prompts, or create another render path.

Ask one short question at a time:

1. Which portrait should I use?
2. Do you want one look or the complete eight-image pack?
3. If one look, which: `mirror-selfie`, `rain-mask`, `car-editorial`,
   `punk-noir`, `art-stop-sign`, `sunroof-flash`, `sword-studio`, or
   `cow-herd-sports-car`?
4. Use Nano Banana 2 by default, Lite for economy, or Pro for premium?

Before paid work:

```bash
npm run format:skai-image -- check --format=cinematic-portrait-pack
npm run format:skai-image -- smoke --format=cinematic-portrait-pack
```

For every selected look, use a unique run id:

```bash
npm run format:skai-image -- init --format=cinematic-portrait-pack --run=<id> --variant=<look> --input=<absolute-image-path> [--model=<route>]
npm run format:skai-image -- validate --format=cinematic-portrait-pack --run=<id>
npm run format:skai-image -- estimate --format=cinematic-portrait-pack --run=<id>
```

Stop and ask once before spending. After explicit approval:

```bash
npm run format:skai-image -- render --format=cinematic-portrait-pack --run=<id> --approve-paid
npm run format:skai-image -- inspect --format=cinematic-portrait-pack --run=<id>
npm run format:skai-image -- inspect --format=cinematic-portrait-pack --run=<id> --visual-pass --review-notes="<specific observations>"
npm run format:skai-image -- finalize --format=cinematic-portrait-pack --run=<id> --approve-final
```

Hard rules:

- Preserve the same face and identity in every selected look.
- Keep the exact packaged prompt unless the user explicitly requests a change.
- State that one look is one paid prediction; the complete pack uses eight.
- Validate every run before spending.
- Never generate without `--approve-paid`.
- Persist and resume each prediction id instead of replacing timed-out jobs.
- Never exceed three attempts for any look.
- View every output and record look-specific notes before finalizing.
- Never print, store, or commit `REPLICATE_API_TOKEN`.

{
  "runtime": {
    "node": ">=22",
    "commands": ["npm"]
  },
  "providers": [
    {
      "name": "Replicate",
      "models": [
        "google/nano-banana-2-lite",
        "google/nano-banana-2",
        "google/nano-banana-pro"
      ]
    }
  ],
  "environment": [
    {
      "name": "REPLICATE_API_TOKEN",
      "secret": true,
      "requiredFor": ["render", "resume"]
    }
  ],
  "notes": [
    "All smoke and validation commands are local.",
    "No proof generation was purchased for this package.",
    "Each selected look is one independent paid prediction.",
    "The complete eight-image pack therefore uses eight predictions.",
    "The estimate command prints the selected route before approval.",
    "Paid generation requires explicit approval and is capped at three attempts per look."
  ]
}

{
  "required": [
    {
      "id": "portrait",
      "type": "image",
      "formats": ["jpg", "png", "webp"],
      "minimumWidth": 512,
      "minimumHeight": 512,
      "maximumBytes": 26214400
    }
  ],
  "question": "Which portrait should I use, and do you want one look or the complete eight-image pack?",
  "variants": [
    "mirror-selfie",
    "rain-mask",
    "car-editorial",
    "punk-noir",
    "art-stop-sign",
    "sunroof-flash",
    "sword-studio",
    "cow-herd-sports-car"
  ],
  "defaults": {
    "model": "nano-banana-2",
    "variant": "mirror-selfie",
    "aspectRatio": "3:4",
    "outputCountPerRun": 1
  },
  "modelRoutes": {
    "economy": "google/nano-banana-2-lite",
    "default": "google/nano-banana-2",
    "premium": "google/nano-banana-pro"
  }
}

{
  "progress": "Prepare -> Validate -> Transform -> Inspect",
  "attemptCapPerLook": 3,
  "stages": [
    {
      "id": "prepare",
      "output": "One run pinned to the portrait, selected look prompt, and model route.",
      "paid": false,
      "approvalRequired": false
    },
    {
      "id": "validate",
      "output": "The portrait and selected prompt contract checked locally.",
      "paid": false,
      "approvalRequired": false
    },
    {
      "id": "transform",
      "output": "One selected look per paid prediction; repeat only for approved variants.",
      "paid": true,
      "approvalRequired": true
    },
    {
      "id": "inspect",
      "output": "Automatic file checks plus an actual visual review for every look.",
      "paid": false,
      "approvalRequired": true
    }
  ]
}

A realistic vertical mirror selfie of the same young man from the reference image, with his face and identity preserved exactly as in the source image. His hair is short, slightly messy, and slightly damp. His appearance from the first reference photo must be strictly preserved without any distortion. He is a bearded man. Focus on facial details, skin texture, and clothing textures. He is wearing a white shirt with the buttons open. There are realistic red marks and traces on his face, neck, chest, and arms. He is wearing a silver chain around his neck and is smoking a cigarette. He slightly tilts his head backward to emphasize the jawline. The photo is taken indoors with atmospheric (moody) lighting. The overall look is natural, realistic, and unprocessed, giving the feeling of a casual mirror selfie. The face must not be altered. The person must look exactly the same as in the original image.

An ultra close-up cinematic portrait should be created under heavy nighttime rainfall; the framing must be a tight close-up covering the head and shoulders (85mm lens feel, f/1.8, natural portrait compression), with the camera at eye level and perfectly aligned with the character; the background should be completely dark but not entirely blacked out, allowing very subtle depth with low-level noise and micro points where raindrops catch light.

The main light source should be soft ambient moonlight coming from above and slightly from the upper left; this light should illuminate the upper parts of the face and gradually fade downward. Additionally, a very subtle low-intensity bounce light from the lower right should prevent the face from being completely lost in darkness. There must be absolutely no streetlight or warm tones. Colors must remain vivid and preserved; the red of the Spider-Man mask should appear natural and saturated, and the image must not be black-and-white or faded.

The character’s face must be 100% identical to the reference image, with hyper-realistic skin (micro pores, wetness, natural skin texture); hair should be wet with a few strands sticking to the forehead. The character must definitely be wearing clothing, and a bare torso must never be visible: only the upper part of the Spider-Man suit (red-blue suit covering the chest and shoulders) should be present as a layered costume piece, ensuring no nudity.

The most critical action must be clearly defined: the character should be in the moment of pulling the Spider-Man mask upward from the head (peeling motion); the mask must not be fully removed, it should be lifted only up to the forehead, revealing the upper part of the face while the lower part remains inside the mask. The mask fabric should appear wet, heavier, and under elastic tension, with finger pressure creating realistic deformation in the fabric.

Rain physics must be highly realistic: heavy raindrops falling at different speeds, some forming thin motion-blur streaks due to long exposure, others appearing as sharp droplets hitting the face and mask; micro splashes should occur on impact, and accumulated water droplets on the face and mask should catch the light.

Tears should flow from the character’s eyes, but not in an exaggerated dramatic crying manner—rather a strong and controlled emotional expression: eyes filled with tears, tears mixing with rain and naturally flowing down the cheeks; mouth neutral and slightly tense, jaw muscles tight, eyebrows slightly furrowed.

The scene should have high contrast without losing detail, with controlled highlight roll-off and preserved shadow detail; slight film grain, very subtle lens bloom, and natural vignette should be applied. No AI artifacts, plastic skin, incorrect anatomy, missing clothing, or style drift should be present. The final result should look like an ultra-realistic, emotionally powerful frame taken from a real cinema film, in billboard-quality resolution.

Hair is black, short, slightly damp, and slightly messy. It must be presented exactly like in the first reference image, without any deformation. In front of us is an avant-garde, cinematic editorial photograph. The subject is sitting inside a black car in the driver’s seat, in a low position with the front door open. One hand confidently rests on the steering wheel, and his body is slightly leaning forward. His head is tilted downward, reflecting a calm, introspective, and emotionally controlled mood.

The subject is wearing a dark-toned, oversized, multi-layered, and textured outfit. He is dressed in wide-cut classic trousers and shiny leather boots. Dark sunglasses hide his eyes, giving him a mysterious and timeless presence.

The scene takes place in an open landscape covered with tall, wild grass, extending toward an overexposed sky, creating a large negative space. The lighting is high-contrast yet controlled, with soft transitions and subtle highlights. The image has a matte cinematic finish with slight film grain, low color saturation, and the depth and sharpness typical of medium-format photography.

The man must never have tattoos. The aesthetic is timeless, minimalist, and fashion-focused, directed with a cinematic approach, captured in ultra-high resolution, 4K quality format.

A RAW black-and-white street portrait generated from the input photo.

VIBE: Harsh noir, artistic rebellion, punk-chic style.

ENVIRONMENT: Plain, light-toned background.

COMPOSITION: Tight close-up shot. The person’s hand makes a bold gesture over the eye using two fingers.

LIGHTING: High-contrast black-and-white with dramatic shadows emphasizing intense eye contact.

STYLE: Black crew-neck T-shirt. Heavy silver rings (weapon-themed), silver bracelet, and a thick pearl necklace with silver details.

EXPRESSION: Harsh, piercing gaze with a defiant attitude.

TECHNICAL: 35mm film aesthetic, heavy Kodak Tri-X grain, high contrast, sharp textures.

Create an ultra-realistic, cinematic image with a slightly dreamy film poster aesthetic (no text included). A young man stands under a STOP sign and leans his shoulder against the metal pole; the shot is taken from a low angle, with the STOP sign positioned slightly behind and above the man’s head, clearly visible against a dark navy night sky; the word “ART” is spray-painted in black over the STOP text.

The face must be 100% identical to the uploaded reference photo (bone structure, eyes, nose, lips, skin texture), with no stylization or modification. The camera looks upward from near ground level; the subject is slightly off-center in the frame, while the STOP sign dominates the upper portion; the shoulder physically touches the pole and this contact is clearly felt; vertical framing with strong poster balance and no distracting foreground objects.

The pose is relaxed, slightly rebellious, and confident; weight shifted to one side, shoulder leaning on the pole, head slightly raised, with a cool and distant gaze.

Clothing: oversized black hoodie with white neon line graphics; the neon lines appear bright and striking due to strong flash effect; the hood sits loosely around the neck; the style clearly reflects hip-hop culture; a thick diamond-detailed chain necklace is worn around the neck and sparkles strongly under the flash (no leather jacket or puffer jacket).

Lighting: night scene with direct flash aesthetic; strong front flash from the camera sharply illuminates the subject, while the background remains dark navy; harsh light falloff and sharp shadows are present; strong specular highlights appear on skin and fabric, with slight reflections on the STOP sign.

Color and grading: dark navy night sky, flash-lit neutral-cool skin tones, high contrast with preserved blacks, slight teal tones in shadows, and a modern street editorial look.

Textures and effects: ultra-detailed skin with visible pores, bright and sharp neon lines, realistic fabric wrinkles, subtle lens smudges/light streaks, soft film grain, and very slight bloom in highlights.

The background is completely clean with an empty navy sky, the STOP sign is clear and readable, and no extra objects are present.

Lens and settings: 28mm lens, f/2.8, flash photography aesthetic, high dynamic range, 8K quality.

Aesthetic: cinematic realism, flash photography, street editorial, hip-hop urban energy, modern film frame.

Create an ultra-realistic, cinematic image. Night scene: two men are seen from an angle looking down into the car through the sunroof; the camera looks from above into the vehicle. The man in the driver’s seat looks directly up at the camera, while a second man in the back seat, partially visible and wearing a black ski mask (balaclava), also looks upward.

The driver’s face must be 100% identical to the uploaded reference photo (bone structure, eyes, nose, lips, skin texture), with no stylization or modification; the masked man’s face must remain completely hidden. The camera is positioned directly above the car roof with a slight diagonal angle, looking down through the sunroof; the framing is tight and vertical, with the sunroof frame forming a natural border, the driver placed in the bottom-right of the frame and the masked man in the top-left.

Lighting follows a night + direct flash aesthetic: a strong front flash from the camera sharply illuminates both characters, while a blue/teal ambient light remains in the interior; warm subtle reflections appear on the skin, with high contrast but without crushed blacks; soft reflections are visible on glass and the car body.

Color grading is teal/cyan dominant with slight magenta accents, creating a cold urban night atmosphere with deep blacks and slightly desaturated mid-tones; very subtle chromatic aberration appears at the edges. The driver maintains a neutral, cool expression with natural posture and direct eye contact; he wears simple dark streetwear (crewneck/sweatshirt), while the masked man wears a black hoodie and balaclava, remaining slightly in shadow with only the eyes visible.

The car interior features dark leather seats with realistic stitching details and slight wear marks; the sunroof frame is sharp, with very light dust/dirt particles on the glass surface and environmental reflections on the roof paint.

Texture and realism are at a high level: visible skin pores and minor imperfections, natural fabric behavior, and no artificial plastic look; slight film grain, fine noise, and subtle lens dirt/smudges create an old and realistic photo feeling. The lens is 35mm equivalent with f/2.8 shallow depth of field (faces sharp, edges slightly soft), high dynamic range, 8K detail, and editorial photography quality.

Aesthetic: cinematic realism, flash photography, night street atmosphere, raw urban feel, high-end editorial look, natural imperfections.

Create an ultra-realistic, cinematic, and minimal studio scene (no text included). The scene is illuminated with a completely dark background and a soft circular spotlight on the ground. A man is kneeling on the floor; his body is not completely collapsed forward—on the contrary, despite his situation, he maintains a dignified and strong posture: shoulders upright, back kept as straight as possible, torso slightly leaning forward but not collapsed; there is a sense of pain but no helplessness, with controlled and resilient body language. His head is tilted downward, but his face is not fully hidden—part of the face is slightly turned toward the camera and partially visible.

The face must be 100% identical to the uploaded reference photo (bone structure, eyes, nose, lips, skin texture), with no stylization; the hairstyle must match the reference image exactly but appear slightly damp and messy. The character is shirtless, with strong and realistic back muscles and anatomical structure clearly visible; he wears loose-fitting black fluid-fabric pants on the lower body, and his feet are completely bare, naturally touching the ground.

A dragon tattoo covers the entire back of the character, designed with strong linear composition; the tattoo has black main outlines and deep red details, with balanced and natural red tones (never neon); lines are sharp and controlled, following the back muscles to enhance volume.

A long sword passes through the character’s back and exits from the chest; the sword goes through the exact center, with correct perspective and physically believable alignment; the metal surface is clean and sharp, with a slightly supernatural but subtle soft glow; entry and exit points are clearly visible, but there must be absolutely no blood, wound detail, or gore.

Lighting is set with a soft but focused spotlight coming from above; the light slightly intensifies at the sword’s entry point, creating elegant reflections on the metal and soft highlights on the character’s back, shoulders, and tattoo; a natural diffused shadow forms on the ground, while the background remains completely black.

The camera is close-up (medium-close / tight framing), with the subject positioned in the center of the frame, focusing on the upper body and sword; lens around 50mm with f/2.8–f/4 depth of field, keeping the character and sword sharp while edges remain slightly soft.

Texture and realism are at a very high level: visible skin pores, slight imperfections, and natural moisture on the skin; realistic fabric wrinkles; detailed metal reflections on the sword; very subtle film grain in the image.

Aesthetic: cinematic realism, dramatic minimalism, dark atmosphere, strong character posture, and deep emotional impact.

Create an ultra-realistic, cinematic image with a dramatic editorial tone (no text included). On a narrow forest road, a man sits on top of a stylish silver sports car that is completely surrounded by a dense herd of cows; the camera is positioned closer (medium-close shot) and clearly shows the man, the car, and the cows immediately around the vehicle in high detail. The face must be 100% identical to the uploaded reference photo (bone structure, eyes, nose, lips, skin texture), with no stylization.

The camera is slightly above eye level and angled toward the car, with tight framing and no empty space; the car and the man are centered, while the cows fill the foreground and midground entirely, and especially the cows in the front must be sharp and highly detailed (fur, eyes, horn details clear, no blur). The man sits naturally and comfortably on the car roof, legs slightly apart, hands relaxed; his expression is calm and confident, and he does not look at the camera, instead gazing thoughtfully into the distance.

Clothing: oversized anthracite-colored Nike logo T-shirt with a distressed texture; loose black sweatpants with thin white side stripes; white and black Nike basketball shoes.

The car is a modern silver sports car (Porsche-like silhouette) with highly reflective metallic paint that cleanly and realistically reflects the environment; it has sharp lines and a premium, flawless appearance. The herd of cows naturally surrounds the vehicle in a dense formation, with brown and white pattern variations, showing a realistic, non-repetitive distribution.

The environment is a narrow asphalt road surrounded by a deep, dark forest in the background; trees create a dense dark texture with slight atmospheric depth but no heavy fog. Lighting is soft and slightly diffused daylight; light coming from above/side highlights the metallic surface of the car, with balanced contrast so the man, car, and animals remain clearly visible; subtle highlights appear on fur and metal surfaces.

The color palette uses natural cinematic tones; greens and browns are slightly desaturated, blacks are controlled and deep, and skin tones are warm-neutral. All surfaces are ultra-detailed (skin, fabric, animal fur, car paint) with no artificial look; slight film grain enhances realism.

Lens: 35mm, f/4–f/5.6 (for wide depth of field), high dynamic range, 8K quality.

Aesthetic: cinematic realism, editorial photography, surreal rural contrast, fusion of luxury and chaos.

{
  "automatic": [
    "Exactly 1 decodable JPG, PNG, or WebP output per run",
    "At least 640x800 pixels",
    "Aspect ratio within 0.02 of 3:4",
    "At least 50 KB"
  ],
  "visual": [
    "The same person remains recognizable without face drift",
    "The selected look matches its exact packaged prompt",
    "Face, hands, clothing, props, and scene geometry are coherent",
    "Lighting, lens behavior, grain, and texture remain photographic",
    "The image reads as a finished cinematic editorial portrait"
  ],
  "finalization": {
    "automaticPassRequired": true,
    "visualPassRequired": true,
    "reviewNotesRequired": true,
    "explicitApprovalFlag": "--approve-final"
  }
}