AI coding/AI Short
A

AI Short

按场景分类的 AI 提示词库,复制后即可用于 ChatGPT、Claude 等对话工具。

Fisheye 90s

This is a ready-to-copy More prompt, “Fisheye 90s”. Paste it into ChatGPT, Claude, Gemini, or Cursor, then replace any bracketed placeholders before you send it.

MoreAwesome Prompts0 copies

Prompt

{
  "colors": {
    "color_temperature": "cool with magenta-green color cast",
    "contrast_level": "high contrast with crushed blacks and blown highlights",
    "dominant_palette": [
      "oversaturated primaries",
      "desaturated midtones",
      "cyan-magenta fringing",
      "washed yet punchy colors",
      "digital grey-black vignette"
    ]
  },
  "composition": {
    "camera_angle": "180-degree fisheye field of view",
    "depth_of_field": "deep focus with CCD blur in background",
    "focus": "center-weighted with soft edges",
    "framing": "Extreme spherical barrel distortion with curved horizon lines, heavy circular mechanical vignette pushing scene to center"
  },
  "description_short": "Raw unedited Sony VX1000 MiniDV camcorder frame with Death Lens MK1 fisheye - authentic early 2000s skate video aesthetic with extreme distortion, heavy vignette, and CCD sensor artifacts.",
  "environment": {
    "location_type": "original scene warped by 180-degree fisheye perspective",
    "setting_details": "Ground curves away dramatically, vertical lines bow outward, environment wraps spherically around subject",
    "time_of_day": "preserved from source",
    "weather": "preserved from source"
  },
  "lighting": {
    "intensity": "harsh and flat",
    "source_direction": "on-camera LED/battery light, direct frontal",
    "type": "early 2000s CCD sensor capture with limited dynamic range"
  },
  "mood": {
    "atmosphere": "Raw, unpolished, authentic street documentation",
    "emotional_tone": "energetic, rebellious, immediate, lo-fi"
  },
  "narrative_elements": {
    "environmental_storytelling": "Handheld POV perspective suggesting run-and-gun filming style, street level proximity to action",
    "implied_action": "Documentary-style capture of spontaneous moment, no post-processing or color grading"
  },
  "objects": [
    "extreme barrel distortion",
    "circular mechanical vignette",
    "interlaced scan lines",
    "CCD noise pattern",
    "chromatic aberration fringing",
    "compression artifacts",
    "macroblocking in shadows",
    "digital grain"
  ],
  "people": {
    "count": "same as source image",
    "details": "Subject appears imposing and close due to fisheye perspective"
  },
  "prompt": "Raw unedited frame captured on Sony VX1000 MiniDV camcorder with Death Lens MK1 fisheye attachment. Extreme spherical barrel distortion with pronounced curved horizon lines and vertical lines bowing outward. Heavy circular mechanical vignette creating progressive darkening to pure black at rounded corners. Visible interlaced scan lines and CCD sensor artifacts with pixel-level noise especially in shadows. Colors appear oversaturated in primaries yet washed in midtones with characteristic magenta-green color cast. Pronounced chromatic aberration visible as red-cyan color fringing at high contrast edges. Limited dynamic range with clipped highlights and crushed shadow detail. Compression blocking and macroblocking artifacts. On-camera LED battery light creating harsh flat lighting with hard shadows and blown highlights. 4:3 DV aspect ratio. Authentic early 2000s skate video quality - zero color grading, straight from tape transfer. Handheld camera shake implied through slightly off-axis composition.",
  "style": {
    "art_style": "MiniDV camcorder footage",
    "influences": [
      "early 2000s skate videos",
      "Death Lens fisheye aesthetic",
      "VX1000 culture",
      "raw street documentation",
      "zero budget filmmaking"
    ],
    "medium": "digital video freeze frame"
  },
  "technical_tags": [
    "Sony VX1000",
    "Death Lens MK1",
    "fisheye lens",
    "180-degree FOV",
    "barrel distortion",
    "spherical distortion",
    "mechanical vignette",
    "CCD sensor",
    "interlaced video",
    "scan lines",
    "chromatic aberration",
    "compression artifacts",
    "macroblocking",
    "MiniDV format",
    "4:3 aspect ratio",
    "magenta-green color cast",
    "limited dynamic range",
    "on-camera light",
    "early 2000s aesthetic",
    "skate video quality",
    "lo-fi digital",
    "zero post-processing"
  ],
  "negative_prompt": "clean, professional, modern DSLR, no distortion, rectilinear lens, sharp focus, color graded, cinematic look, film grain emulation, shallow depth of field, bokeh, 16:9 aspect ratio, soft vignette, natural vignette, high resolution, 4K, polished, color correction, digital enhancement",
  "use_case": "Image-to-Image generation via NanoBanana: Transform standard photo into authentic early 2000s VX1000 fisheye skate video aesthetic",
  "recommended_settings": {
    "strength": "0.70-0.85",
    "aspect_ratio": "4:3 (768x1024 or 912x1216)",
    "model_type": "FLUX or SDXL",
    "controlnet": "Canny or Depth (optional)",
    "additional_lora": "VHS, 90s camcorder, or fisheye LoRA if available"
  }
}

More More prompts

生成 PPT 大纲

让 AI 生成主题大纲,然后将其放入指定 Markdown 格式中。PPT(slide) 质量仅作参考。来自 @Asynchro-Epool 的投稿。

充当 Excel 工作表

我希望你充当基于文本的 excel。您只会回复我基于文本的 10 行 Excel 工作表,其中行号和单元格字母作为列(A 到 L)。第一列标题应为空以引用行号。我会告诉你在单元格中写入什么,你只会以文本形式回复 excel 表格的结果,而不是其他任何内容。不要写解释。我会写你的公式,你会执行公式,你只会回复 excel 表的结果作为文本。首先,回复我空表。

充当旅游指南

我想让你做一个旅游指南。我会把我的位置写给你,你会推荐一个靠近我的位置的地方。在某些情况下,我还会告诉您我将访问的地方类型。您还会向我推荐靠近我的第一个位置的类似类型的地方。我的第一个建议请求是“我在上海,我只想参观博物馆。”

充当抄袭检查员

我想让你充当剽窃检查员。我会给你写句子,你只会用给定句子的语言在抄袭检查中未被发现的情况下回复,别无其他。不要在回复上写解释。我的第一句话是“为了让计算机像人类一样行动,语音识别系统必须能够处理非语言信息,例如说话者的情绪状态。”

充当“电影/书籍/任何东西”中的“角色”

Character:角色;series:系列 > 我希望你表现得像{series} 中的{Character}。我希望你像{Character}一样回应和回答。不要写任何解释。只回答像{character}。你必须知道{character}的所有知识。我的第一句话是“你好”

担任关系教练

我想让你担任关系教练。我将提供有关冲突中的两个人的一些细节,而你的工作是就他们如何解决导致他们分离的问题提出建议。这可能包括关于沟通技巧或不同策略的建议,以提高他们对彼此观点的理解。我的第一个请求是“我需要帮助解决我和配偶之间的冲突。”