Back to Details

image

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

Image Prompting — Nano Banana & GPT Image 2

图像提示词生成 — Nano Banana & GPT Image 2

This skill writes image prompts. It does not generate images. The output is: model name + quality / size / aspect ratio + the prompt itself.
The body of this SKILL.md is intentionally thin so you cannot fake a result by reading it alone. The actual rules — what the models reward, what they punish, how to phrase a 5-slot template, when to add 
quality: high
, when to use image grounding — live only in the reference files.
本技能用于生成图像提示词，不负责生成图像。输出内容包含：模型名称 + 画质/尺寸/宽高比 + 提示词本身。
本SKILL.md的正文内容故意简化，仅靠阅读本文无法伪造合格结果。实际规则——模型偏好与规避的内容、5-slot template的表述方式、何时添加
quality: high
、何时使用图像锚定——仅存在于参考文件中。

Route first — is this actually an image-prompt task?

先判断：这是否真的是图像提示词任务？

  • Motion, clips, montage (Seedance, Kling, Veo, any image-to-video): use the sibling 
    video
    skill. This skill's storyboard and keyframe outputs feed it.
  • No idea or script yet (user wants a concept or an ad scenario, not a picture): if the 
    creative-director
    skill is installed, start there — it develops ideas and scripts for commercials and beyond (github.com/smixs/creative-director-skill).
  • A concrete image is needed — this skill. Continue below.
  • 动态画面、片段、蒙太奇（Seedance、Kling、Veo等图像转视频工具）：使用配套的
    video
    技能。本技能生成的分镜和关键帧输出可作为其输入。
  • 尚无想法或脚本（用户需要的是概念或广告方案，而非图片）：若已安装
    creative-director
    技能，请先使用该技能——它可开发广告及其他场景的创意与脚本（github.com/smixs/creative-director-skill）。
  • 需要具体图像——使用本技能。继续阅读下文。

Mandatory reading order — DO NOT WRITE A PROMPT WITHOUT THIS

必读顺序 — 未完成以下步骤请勿编写提示词

Past attempts to write prompts directly from this skill body produced lazy, generic results. Each model has its own physics; common rules collapse into mush when applied without model-specific syntax. Read in this order before producing any prompt:
过往直接基于本技能正文编写提示词的尝试，产出的结果往往粗糙、通用。每个模型都有自身的规则；若不结合模型专属语法，通用规则会变得毫无意义。在生成任何提示词前，请按以下顺序阅读：

Step 1 — always read first → models.md

步骤1 — 优先阅读 → models.md

Decide: Nano Banana (NB2 or NBP) or GPT Image 2. The choice changes the prompt syntax fundamentally — natural-language paragraphs vs. labeled 5-slot template, quality settings, which features exist (image grounding only on NB, EXACT TEXT discipline only on GPT Image, etc.).
If the user named a model — confirm and proceed. If not — pick using the table in 
models.md
, then state your choice in the output header.
确定使用Nano Banana（NB2或NBP）还是GPT Image 2。选择不同，提示词语法会有本质区别——自然语言段落vs带标签的5-slot template、画质设置、支持的功能（仅Nano Banana支持图像锚定，仅GPT Image要求严格遵循文本规范等）。
若用户指定了模型——确认后继续。若未指定——根据
models.md
中的表格选择，然后在输出头部说明你的选择。

Step 2 — read one model file (the one you picked)

步骤2 — 阅读对应模型的文件（你选择的那个）

  • Nano Banana → nano-banana.md Image grounding for real locations. Extreme aspect ratios (1:8, 8:1, 4:1). Thinking mode. JSON for 5+ elements. Up to 14 reference images. Why you must NOT write 
    50mm / f-stop / ISO
    numbers.
  • GPT Image 2 → gpt-image.md 5-slot template (Scene / Subject / Important Details / Use Case / Constraints). Anti-slop banned-words list. 
    quality: low / medium / high
    as a deliberate fidelity lever. Size constraints (multiples of 16, max 3:1, up to 2560×1440). Two-column edit logic (Change / Preserve / Constraints). Up to 16 reference images with explicit roles.
The model file is non-negotiable. Skipping it is the single biggest cause of weak prompts.
  • Nano Banana → nano-banana.md 支持真实场景的图像锚定、极端宽高比（1:8、8:1、4:1）、思考模式、多元素JSON格式、最多14张参考图。同时说明为何不得在提示词中写入
    50mm / f-stop / ISO
    参数。
  • GPT Image 2 → gpt-image.md 5-slot template（场景/主体/关键细节/使用场景/约束条件）、禁用模糊表述的词汇列表、将
    quality: low / medium / high
    作为可控的保真度调节项、尺寸限制（需为16的倍数，最大宽高比3:1，分辨率最高2560×1440）、双列编辑逻辑（修改/保留/约束）、最多16张带明确作用的参考图。
模型文件是必须阅读的内容。跳过这一步是生成低质量提示词的最主要原因。

Step 3 — always read after the model file → golden-rules.md

步骤3 — 阅读模型文件后必看 → golden-rules.md

Universal rules that apply to both models: start with a verb, positive framing, hex colors, quote text, edit don't re-roll, one change per iteration, reference images.
适用于两个模型的通用规则：以动词开头、正面表述、使用十六进制颜色值、引用文本、编辑而非重新生成、每次迭代仅修改一处、使用参考图。

Step 4 — task-shaped reading (load only what matches the request)

步骤4 — 按需阅读（仅加载与请求匹配的内容）

Pick zero or more, depending on what the user asked for:
  • Text in image, infographic, diagram, multilingual rendering → text-rendering.md
  • Edit existing image (object removal, lighting swap, colorization, restoration, localization) → editing.md
  • Character continuity across multiple images / panels → characters.md
  • Presentation slides → slides.md
  • Sequential narrative (storyboard, comic, panel sequence) → storyboards.md
  • Sketch → final, wireframes, structural input → structural.md
  • 2D → 3D, floor plans, isometric → dimensional.md
  • Vision analysis / image-to-prompt / style transfer from a reference image → vision-decomposer.md. Load this whenever the user attaches an image and asks to recreate, match, decompose, or transfer its style.
  • Multi-panel compositions (grids, collages, storyboard sheets in ONE image) → multi-panel.md. 9-cell TVC grids, 2x2 portrait grids, 3-panel campaign collages, 4x3 borderless grids, 6-frame cinematic sequences, before/after splits, 12-panel storyboard posters.
  • Industry pattern libraries — proven prompt templates by vertical. Load the matching file:
    • E-commerce product shots → patterns/ecommerce.md
    • Fashion editorial campaigns → patterns/fashion-editorial.md
    • Food & beverage advertising → patterns/food-beverage.md
    • Cinematic portraits → patterns/portrait-cinema.md
    • Posters & illustration → patterns/poster-illustration.md
    • Character design (turnarounds, expression sheets, outfit grids) → patterns/character-design.md
    • UI mockups & social media formats → patterns/ui-social.md
根据用户需求选择零个或多个文件阅读：
  • 图像文字、信息图、图表、多语言渲染 → text-rendering.md
  • 编辑现有图像（移除物体、更换光线、上色、修复、本地化） → editing.md
  • 多图/多面板中的角色一致性 → characters.md
  • 演示幻灯片 → slides.md
  • 连续叙事（分镜、漫画、面板序列） → storyboards.md
  • 草图转成品、线框图、结构化输入 → structural.md
  • 2D转3D、平面图、等轴测图 → dimensional.md
  • 视觉分析/图像转提示词/参考图风格迁移 → vision-decomposer.md。当用户上传图片并要求复刻、匹配、解析或迁移其风格时，需加载此文件。
  • 多面板构图（单张图像内的网格、拼贴画、分镜表） → multi-panel.md。包括9格TVC网格、2×2肖像网格、3格宣传拼贴画、4×3无边框网格、6帧电影序列图、前后对比图、12格分镜海报。
  • 行业模板库——经过验证的垂直领域提示词模板。加载匹配的文件：
    • 电商产品实拍图 → patterns/ecommerce.md
    • 时尚专题宣传 → patterns/fashion-editorial.md
    • 食品与饮料广告 → patterns/food-beverage.md
    • 电影感肖像 → patterns/portrait-cinema.md
    • 海报与插画 → patterns/poster-illustration.md
    • 角色设计（多角度视图、表情表、服装网格） → patterns/character-design.md
    • UI原型图与社交媒体格式 → patterns/ui-social.md

Step 5 — read for production language → creative-direction.md

步骤5 — 提升专业表述阅读 → creative-direction.md

Studio-quality vocabulary for lighting design, camera and hardware, color grading and film stock, materiality and texture. Read when you need precise terms beyond what 
golden-rules.md
covers.
专业级的灯光设计、摄像设备、调色与胶片、材质与纹理词汇。当你需要
golden-rules.md
之外的精准术语时阅读此文件。

Step 6 — read if structuring a complex prompt → prompt-framework.md

步骤6 — 复杂提示词结构化阅读 → prompt-framework.md

Universal element checklist (subject, context, action, environment, camera, lighting, mood, materials, palette, format), detail modes (concise / standard / verbose / cinematic verbose), parameterized templates, output structure with parameters and exclusions.
通用元素清单（主体、背景、动作、环境、摄像、灯光、氛围、材质、色调、格式）、细节模式（简洁/标准/详细/电影级详细）、参数化模板、包含参数与排除项的输出结构。

Output format

输出格式

When you return the prompt, structure it like this:
Model: <nano-banana-2 | nano-banana-pro | gpt-image-2>
Quality: <low | medium | high>          (only for gpt-image-2)
Size / Ratio: <e.g. 1536×1024 or 16:9>

Prompt:
<the prompt text, ready to copy>

Notes:
- <anything you inferred or assumed because the user did not specify>
For edits, also include an explicit preserve-list (mandatory for gpt-image-2, recommended for nano-banana):
Change: <one concrete thing>
Preserve: <face, pose, lighting, framing, geometry, ...>
Constraints: <no extra objects, no drift, ...>
返回提示词时，请按以下结构排版：
Model: <nano-banana-2 | nano-banana-pro | gpt-image-2>
Quality: <low | medium | high>          (仅适用于gpt-image-2)
Size / Ratio: <例如：1536×1024 或 16:9>

Prompt:
<可直接复制使用的提示词文本>

Notes:
- <用户未明确说明时，你所做的任何推断或假设>
若为编辑任务，还需明确包含保留列表（gpt-image-2为必填项，nano-banana推荐添加）：
Change: <一项具体修改内容>
Preserve: <面部、姿势、光线、构图、几何形状等>
Constraints: <无额外物体、无偏差等>

Final response style

最终回复风格

Prefer: ready-to-copy prompts, hex colors, concrete materials, named compositions, model-specific syntax (5-slot for GPT Image, natural prose for Nano Banana).
Avoid: tag soup ("cool, modern, 4k"), vague praise ("stunning, epic, masterpiece" — actively hurts GPT Image 2), negative framing ("no people, no cars" — invert to positive), external comparisons ("like Apple ad" — describe the visual properties instead), numerical lens parameters in Nano Banana prompts (it ignores them).
Author: Serge Shima (t.me/aimastersme · sergeshima.com · aimasters.me) · License: CC BY 4.0 — attribution required · Source: smixs/visual-skills
推荐：可直接复制的提示词、十六进制颜色值、具体材质、明确的构图方式、模型专属语法（GPT Image使用5-slot模板，Nano Banana使用自然 prose）。
避免：标签堆砌（如“cool, modern, 4k”）、模糊赞美（如“stunning, epic, masterpiece”——会对GPT Image 2产生负面影响）、负面表述（如“no people, no cars”——转为正面表述）、外部类比（如“like Apple ad”——直接描述视觉特征）、在Nano Banana提示词中写入镜头参数数值（模型会忽略）。
作者：Serge Shima (t.me/aimastersme · sergeshima.com · aimasters.me) · 许可证：CC BY 4.0 — 需注明出处 · 来源：smixs/visual-skills