jeffstric/zjt/location-prop-image-designer

IF 场景数量 >= 2:
    尝试调用 generate_4grid_location_images
    IF 场景数量 < 4:
        用全高黑色占位图片补齐到4个（提示词："A pure black background image, solid black color, 16:9 aspect ratio, full height"）
        顺序：先填充上方两个位置（Shot 1, Shot 2），再填充下方两个位置（Shot 3, Shot 4）
    ENDIF
    IF 4宫格生成失败:
        记录失败原因
        改用 generate_text_to_image 逐个生成
    ENDIF
ELSE IF 场景数量 == 1:
    调用 generate_text_to_image
ENDIF
 
IF 道具数量 >= 2:
    尝试调用 generate_4grid_prop_images
    IF 道具数量 < 4:
        用全高黑色占位图片补齐到4个（提示词："A pure black background image, solid black color, 16:9 aspect ratio, full height"）
        顺序：先填充上方两个位置（Shot 1, Shot 2），再填充下方两个位置（Shot 3, Shot 4）
    ENDIF
    IF 4宫格生成失败:
        记录失败原因
        改用 generate_text_to_image 逐个生成
    ENDIF
ELSE IF 道具数量 == 1:
    调用 generate_text_to_image
ENDIF

{
  "grid_layout": "2x2",
  "grid_aspect_ratio": "16:9",
  "global_watermark": "",
  "shots": [
    {
      "shot_number": "Shot 1",
      "prompt_text": "场景/道具1的完整提示词"
    },
    {
      "shot_number": "Shot 2",
      "prompt_text": "场景/道具2的完整提示词"
    },
    {
      "shot_number": "Shot 3",
      "prompt_text": "场景/道具3的完整提示词"
    },
    {
      "shot_number": "Shot 4",
      "prompt_text": "场景/道具4的完整提示词"
    }
  ]
}

┌─────────┬─────────┐
│ Shot 1  │ Shot 2  │  场景/道具1(左上) → 场景/道具2(右上)
│ (左上)  │ (右上)  │
├─────────┼─────────┤
│ Shot 3  │ Shot 4  │  场景/道具3(左下) → 场景/道具4(右下)
│ (左下)  │ (右下)  │
└─────────┴─────────┘

ABSOLUTELY NO text, NO watermark, NO letters, NO characters, NO words, NO signs, NO writing, NO typography, NO labels, NO captions, NO subtitles, NO Chinese characters, NO English text, NO numbers, NO logos, NO stamps, NO seals, completely text-free image, pure visual content without any written language

{
  "success": true,
  "task_id": 26,
  "name": "GPT Image 2 图片编辑",
  "computing_power": 2,
  "supported_sizes": ["1k", "2k", "4k"],
  "supported_ratios": ["1:1", "2:3", "3:2", "16:9", "9:16"],
  "supports_grid_image": true,
  "default_size": "1k",
  "max_size": "4k",
  "cost_per_image_default_size": 2,
  "cost_per_image_max_size": 2,
  "driver_hint": "提示词长度请控制在5000字符以内，超出将被API拒绝。请精简描述，避免冗余。"
}

1. 调用 read_world() 获取 visual_style
2. 检查 visual_style 中是否包含"写实"、"真实"、"照片"、"摄影"等关键词
3. 如果包含 → 绝对禁止在提示词中使用以下词汇：
   ❌ "anime"
   ❌ "reference sheet"
   ❌ "design reference"
   ❌ "illustration"
   
   必须使用以下词汇：
   ✅ "photography"
   ✅ "photograph"
   ✅ "architectural photography"
   ✅ "location shoot"

[画风描述], [时代环境], [色彩基调]. A professional architectural photography of [场景英文名] ([场景中文名]) from [作品/世界名称], shot with high-quality camera equipment. The location is captured without any people or objects present.
 
Wide-angle photograph showing the complete overview of [场景名] featuring [详细的环境描述，包括地形、建筑、植被、天气]. The scene displays [空间布局描述，包括主要区域、地标、道路]. Architectural style: [建筑风格详述]. Environmental features: [环境特征如光照、氛围、季节]. The location is completely empty with no people or objects. The overall mood conveys [基于场景特点的氛围描述]. Natural lighting and realistic textures are emphasized. No text, signs, or written language appear in the scene. ABSOLUTELY NO text, NO watermark, NO letters, NO characters, NO words, NO signs, NO writing, NO typography, NO labels, NO captions, NO subtitles, NO Chinese characters, NO English text, NO numbers, NO logos, NO stamps, NO seals, completely text-free image, pure visual content without any written language.

[画风描述], [时代环境], [色彩基调]. A professional location design reference sheet for [场景英文名] ([场景中文名]) from [作品/世界名称], set on a clean neutral background. The layout shows the location without any characters or props. Avoid including any text, words, or written characters in the image.
 
Large detailed illustration showing the complete overview of [场景名] featuring [详细的环境描述，包括地形、建筑、植被、天气]. The scene displays [空间布局描述，包括主要区域、地标、道路]. Architectural style: [建筑风格详述]. Environmental features: [环境特征如光照、氛围、季节]. The location is completely empty with no people or objects. The overall mood conveys [基于场景特点的氛围描述]. Do not include any text, signs, or written language in the scene. ABSOLUTELY NO text, NO watermark, NO letters, NO characters, NO words, NO signs, NO writing, NO typography, NO labels, NO captions, NO subtitles, NO Chinese characters, NO English text, NO numbers, NO logos, NO stamps, NO seals, completely text-free image, pure visual content without any written language.

[画风描述], [时代环境], [色彩基调]. A professional product photography collection of [道具英文名] ([道具中文名]) from [作品/世界名称], shot in studio lighting against a clean white backdrop. The prop is photographed alone without any hands, characters, or other objects.
 
Three-Angle Photography: Three high-quality photographs of [道具名] arranged horizontally. Front Photograph (Left): [道具正面照片的详细描述，包括形状、材质、颜色、纹理、装饰细节]. Side Photograph (Center): [道具侧面照片的详细描述，展现厚度、轮廓、连接部分]. Back Photograph (Right): [道具背面照片的详细描述，必须与正面保持一致的设计元素和比例]. 
 
Physical characteristics: [尺寸比例描述], [材质质感如金属光泽、木纹理、宝石透明度等], [表面装饰如雕刻、镶嵌、符文等], [磨损状态和使用痕迹]. All three photographs maintain perfect consistency in proportions, colors, materials, and decorative elements. The prop shows [基于功能和历史的整体状态描述]. Professional studio lighting highlights textures and details. No characters, hands, text, words, or other objects appear in any photograph. ABSOLUTELY NO text, NO watermark, NO letters, NO characters, NO words, NO signs, NO writing, NO typography, NO labels, NO captions, NO subtitles, NO Chinese characters, NO English text, NO numbers, NO logos, NO stamps, NO seals, completely text-free image, pure visual content without any written language.

[画风描述], [时代环境], [色彩基调]. A clean prop design reference sheet for [道具英文名] ([道具中文名]) from [作品/世界名称], set on a white background. The prop appears alone without any characters or hands holding it. Avoid including any text, words, or written characters in the image.