Phantom Strike

{
  "title": "Phantom Strike",
  "description": "An intense, high-octane action shot of a lone warrior battling supernatural entities in a decayed industrial setting.",
  "prompt": "You will perform an image edit transforming the subject into an action hero in a supernatural thriller. The image must be photorealistic, highly detailed, and emulate a frame shot on Arri Alexa with cinematic lighting and a shallow depth of field. The scene depicts the female subject in a derelict, flooded subway tunnel, engaged in mortal combat. She is fighting with shadows that seem to manifest as physical, smoky tendrils extending from the darkness. The lighting is dramatic, highlighting the texture of her skin and the splashing water.",
  "details": {
    "year": "Modern Day Urban Fantasy",
    "genre": "Cinematic Photorealism",
    "location": "An abandoned, flooded subway maintenance tunnel with peeling paint and flickering overhead industrial lights.",
    "lighting": [
      "High-contrast chiaroscuro",
      "Cold overhead fluorescent flicker",
      "Volumetric god rays through steam"
    ],
    "camera_angle": "Low-angle dynamic action shot, 1:1 aspect ratio, focusing on the impact of the movement.",
    "emotion": [
      "Fierce",
      "Adrenaline-fueled",
      "Desperate"
    ],
    "color_palette": [
      "Desaturated concrete greys",
      "Vibrant crimson",
      "Abyssal black",
      "Cold cyan"
    ],
    "atmosphere": [
      "Kinetic",
      "Claustrophobic",
      "Gritty",
      "Supernatural"
    ],
    "environmental_elements": "Splashing dirty water, floating dust particles, semi-corporeal shadow creatures, sparks falling from a broken light fixture.",
    "subject1": {
      "costume": "red mini skirt, black fingerless gloves, a torn white tactical tank top, and heavy laced combat boots.",
      "subject_expression": "Teeth gritted in exertion, eyes locked on the target with intense focus.",
      "subject_action": "fighting with shadows"
    },
    "negative_prompt": {
      "exclude_visuals": [
        "sunlight",
        "blue skies",
        "static poses",
        "smiling",
        "cleanliness"
      ],
      "exclude_styles": [
        "cartoon",
        "anime",
        "3D render",
        "oil painting",
        "sketch"
      ],
      "exclude_colors": [
        "pastel pink",
        "warm orange",
        "spring green"
      ],
      "exclude_objects": [
        "guns",
        "swords",
        "modern vehicles",
        "bystanders"
      ]
    }
  }
}

Run this prompt on a real model without leaving the page. Every run is saved to your history for this prompt.

How to use

1. 1Copy the prompt as-is — no variables required.
2. 2Paste it into ChatGPT, Claude, Gemini, or any capable model.
3. 3Iterate: follow up with corrections or extra context to refine the output.