nano-banana

分类: 内容与多媒体 | 上传者: devonjones | 下载: 0 | 版本: v1.0（最新）

使用 Google 的 Gemini 3 Pro 图像模型 (Nano Banana Pro) 生成、编辑和合成图像。当用户要求创建图像、生成视觉效果、编辑照片、合成多个图像、创建徽标、缩略图、信息图表、产品照片或任何图像生成任务时，请使用此技能。支持文本到图像、图像编辑、多图像合成（最多 14 个图像）、迭代细化、长宽比控制以及基于 Google 搜索的图像生成，以实现实时数据可视化。

更新日志: Source: GitHub https://github.com/devonjones/devon-claude-skills

目录结构

当前层级: tree/main/plugins/nano-banana/skills/nano-banana/

📁 scripts/
- 📄 chat_image.py 3.1 KB
- 📄 compose_images.py 3.6 KB
- 📄 edit_image.py 2.9 KB
- 📄 generate_image.py 2.6 KB
- 📄 search_grounded_image.py 3.0 KB
📄 requirements.txt 27 B
📄 SKILL.md 5.2 KB

SKILL.md

---
name: nano-banana
description: Generate, edit, and compose images using Google's Gemini 3 Pro Image model (Nano Banana Pro). Use this skill when the user asks to create images, generate visuals, edit photos, compose multiple images, create logos, thumbnails, infographics, product shots, or any image generation task. Supports text-to-image, image editing, multi-image composition (up to 14 images), iterative refinement, aspect ratio control, and Google Search-grounded image generation for real-time data visualization.
---

# Nano Banana Pro

Image generation skill powered by Google's Gemini 3 Pro Image model. Enables text-to-image generation, image editing, multi-image composition, and real-time data visualization.

## Requirements

- `GEMINI_API_KEY` environment variable set
- Python packages: `google-genai`, `Pillow`

Install dependencies:
```bash
pip install -r requirements.txt
```

## Quick Start

### Generate an Image

```bash
python scripts/generate_image.py "A cat wearing a wizard hat" cat.png
```

### Edit an Existing Image

```bash
python scripts/edit_image.py photo.png "Add a sunset to the background" edited.png
```

### Compose Multiple Images

```bash
python scripts/compose_images.py "Create a group photo in an office" team.png person1.png person2.png
```

## Available Scripts

| Script | Purpose |
|--------|---------|
| `generate_image.py` | Text-to-image generation |
| `edit_image.py` | Edit/modify existing images |
| `compose_images.py` | Combine up to 14 reference images |
| `chat_image.py` | Interactive multi-turn refinement |
| `search_grounded_image.py` | Generate images with real-time search data |

## Generation Options

### Aspect Ratios
`1:1`, `2:3`, `3:2`, `3:4`, `4:3`, `4:5`, `5:4`, `9:16`, `16:9`, `21:9`

### Resolutions
`1K` (1024px), `2K`, `4K`

### Usage with Options

```bash
python scripts/generate_image.py "Futuristic motorcycle on Mars" mars.png --aspect 16:9 --size 4K
```

## Task Workflows

### Logo Generation

```bash
python scripts/generate_image.py "Clean black-and-white logo with text 'Daily Grind', sans-serif font, coffee bean icon, minimalist style" logo.png --aspect 1:1
```

### Product Mockup

```bash
python scripts/generate_image.py "Studio-lit product photo on polished concrete, 3-point softbox, 45-degree angle, professional e-commerce style" product.png --aspect 4:3 --size 4K
```

### Photorealistic Portrait

```bash
python scripts/generate_image.py "A photorealistic close-up portrait, shot on 85mm lens, golden hour lighting, shallow depth of field, cinematic" portrait.png --size 4K
```

### Stylized Art (Anime/Sticker)

```bash
python scripts/generate_image.py "A kawaii red panda sticker, bold outlines, cel-shading, white background, cute expression" sticker.png
```

### Iterative Design Refinement

Use the chat script for back-and-forth refinement:

```bash
python scripts/chat_image.py
```

Then interact:
```
> Create a logo for 'Acme Corp'
[Image generated]
> Make the text bolder and add a blue gradient
[Refined image]
> save acme_logo.png
```

### Real-Time Data Visualization

Generate infographics with current data:

```bash
python scripts/search_grounded_image.py "Visualize today's weather in Tokyo as an infographic" tokyo_weather.png --aspect 9:16
```

Use cases:
- Live stock-market infographics
- Breaking-news visuals
- Weather dashboards
- Current event visualizations

### Multi-Image Composition

Combine reference images:

```bash
python scripts/compose_images.py "Create a product comparison shot with these items side by side, professional lighting" comparison.png item1.png item2.png item3.png --aspect 16:9
```

Use cases:
- Product comparison shots
- Character sheets
- Team photos
- Style-consistent image series

## Inline Python Usage

For integration in larger scripts:

```python
import os
from google import genai
from google.genai import types

client = genai.Client(api_key=os.environ["GEMINI_API_KEY"])

response = client.models.generate_content(
    model="gemini-3-pro-image-preview",
    contents=["A serene mountain landscape at dawn"],
    config=types.GenerateContentConfig(
        response_modalities=['TEXT', 'IMAGE'],
        image_config=types.ImageConfig(
            aspect_ratio="16:9",
            image_size="2K"
        )
    )
)

for part in response.parts:
    if part.inline_data:
        image = part.as_image()
        image.save("landscape.png")
```

### Editing with Inline Code

```python
from PIL import Image
from google import genai
from google.genai import types

client = genai.Client(api_key=os.environ["GEMINI_API_KEY"])
img = Image.open("input.png")

response = client.models.generate_content(
    model="gemini-3-pro-image-preview",
    contents=["Add dramatic clouds to the sky", img],
    config=types.GenerateContentConfig(
        response_modalities=['TEXT', 'IMAGE']
    )
)
```

## Prompting Tips

**Photorealistic**: Include camera settings, lighting, lens details
```
"Shot on 85mm lens, golden hour lighting, shallow depth of field"
```

**Logos**: Specify style, colors, typography
```
"Clean minimalist logo, sans-serif font, monochrome, vector style"
```

**Product shots**: Describe studio setup
```
"Studio-lit, 3-point softbox, polished surface, 45-degree angle"
```

**Stylized art**: Name the style explicitly
```
"Anime style, cel-shading, bold outlines, vibrant colors"
```

登录后下载/点赞/收藏 ❤ 6 | ★ 0

nano-banana

目录结构

SKILL.md

举报内容

提示