Banana Pro Image Generation

Banana Pro Image Generation Skill

Use this skill when the user needs to generate images, design logos, create whiteboard diagrams, or make social media graphics.

Core Capabilities

You can help users with:

Text-to-image: Generate images in various styles based on text descriptions
Whiteboard diagrams: Generate handwritten-style concept diagrams, flowcharts, and comparison tables
Logo design: Create minimalist, modern logos and icons
Social media graphics: Generate graphics suitable for various platforms
Concept visualization: Transform abstract concepts into intuitive images

Use Cases

Scenario 1: The user needs a tutorial illustration

The user says: "Help me generate a whiteboard diagram showing the core architecture of OpenClaw"

You should：

Understand the user's needs (whiteboard diagram, handwritten style, architecture diagram)
Build a detailed prompt
Call the generation script
Return the generated image

Example prompt：

生成一张白板图片，手写字体风格，内容是OpenClaw核心架构：

标题：OpenClaw 核心架构

三大组件：
📦 Gateway 网关
• 会话管理
• 消息路由
• 多平台连接

🤖 AI 智能体
• 理解指令
• 执行任务
• 返回结果

🔧 Skills 系统
• 扩展功能
• 无限可能

工作流程（用箭头连接）：
用户消息 → Gateway → AI智能体 → Skills → 返回结果

用手写字体，添加框图、箭头、图标等手绘元素

Scenario 2: The user needs a logo design

The user says: "Help me design a logo for an AI assistant"

You should：

Ask about specific requirements (color, style, purpose)
Build a professional logo design prompt
Generate multiple options for selection

Example prompt：

设计一个AI助手Logo，要求：
- 形状：圆形或方形
- 颜色：蓝色渐变
- 元素：抽象的大脑或机器人图案
- 风格：现代简约，科技感
- 用途：适合用在社交媒体头像和网站

Scenario 3: The user needs a comparison chart

The user says: "Generate a comparison chart showing OpenClaw vs ChatGPT"

You should：

Organize the comparison dimensions
Build a clear table-style prompt
Generate an easy-to-read comparison chart

Example prompt：

生成一张对比表格图片，清晰易读，内容是：

标题：OpenClaw vs ChatGPT

对比维度：
1. 部署方式
   OpenClaw: 本地/云端 ✓
   ChatGPT: 在线服务

2. 文件访问
   OpenClaw: 支持 ✓
   ChatGPT: 不支持 ✗

3. 功能扩展
   OpenClaw: Skills系统 ✓
   ChatGPT: 固定功能 ✗

4. 成本
   OpenClaw: 按需付费 ✓
   ChatGPT: 订阅制

使用表格形式，清晰对比

How to Call

Basic call

python scripts/generate_image.py \
  --prompt "你构建的详细提示词" \
  --filename "output.png"

High-definition images (2K/4K)

python scripts/generate_image.py \
  --prompt "你的提示词" \
  --filename "output.png" \
  --resolution 2K

Configure the API using environment variables

export NEXTAI_API_KEY="用户的API密钥"
python scripts/generate_image.py --prompt "你的提示词"

Prompt Construction Techniques

Technique 1: Whiteboard diagram prompt template

生成一张白板图片，手写字体风格，内容包含：
- 标题：[主题]
- 核心要点：
  1. [要点1]
  2. [要点2]
  3. [要点3]
- 使用箭头、框图、列表等手绘元素

Technique 2: Logo design prompt template

设计一个[主题]Logo，要求：
- 形状：[圆形/方形/抽象]
- 颜色：[主色调]
- 元素：[核心元素]
- 风格：[现代/复古/简约]
- 用途：[社交媒体/网站/名片]

Technique 3: Comparison chart prompt template

生成一张对比图，展示[A vs B]：
- 对比维度1：[A的特点] vs [B的特点]
- 对比维度2：[A的特点] vs [B的特点]
- 对比维度3：[A的特点] vs [B的特点]
使用表格或左右对比的形式，清晰易读

Technique 4: Concept diagram prompt template

生成一张概念图，展示[主题]：
- 中心概念：[核心]
- 相关概念：[概念1、概念2、概念3]
- 概念之间的关系用箭头连接
使用思维导图或网络图形式

Workflow

When the user requests image generation, you should:

Understand the requirements
- Image type (whiteboard diagram/logo/graphic/comparison chart)
- Style requirements (handwritten/modern/minimalist)
- Use scenario (tutorial/social media/presentation)
Build the prompt
- Use detailed descriptions
- Specify style and elements
- Add composition suggestions
Call generation
- Choose an appropriate resolution
- Set the output filename
- Execute the generation command
Return the result
- Display the generated image
- Explain the image's features
- Provide revision suggestions

Notes

Prompts should be detailed: The more detailed the description, the better the generation result
Specify the style: Clearly state the required style (handwritten/modern/minimalist, etc.)
Reasonable resolution：
- Daily use: 1K (default)
- HD display: 2K
- Print output: 4K
Cost control：
- 1K: ~$0.04/image
- 2K: ~$0.08/image
- 4K: ~$0.16/image

Practical examples

Example 1: Tutorial illustration

User: "Help me generate a summary image for Chapter 1"

Your reply：

好的，我来为第1章生成一张总结白板图。

[调用生成脚本]
python scripts/generate_image.py \
  --prompt "生成一张白板图片，手写字体风格，内容是OpenClaw第1章核心要点总结：

标题：OpenClaw 是什么？

核心定义：
• AI智能体Gateway网关
• 完全本地部署
• 数据隐私有保障

4大核心优势：
1️⃣ 本地部署 - 保护隐私
2️⃣ 文件访问 - 智能搜索
3️⃣ Skills扩展 - 无限可能
4️⃣ 多平台支持 - 随时随地

用手写字体，添加箭头、框图、下划线等手绘元素" \
  --filename "chapter1_summary.png"

✅ 已生成第1章总结图：chapter1_summary.png

Example 2: Logo design

User: "Design a logo for a super individual"

Your reply：

好的，我来为"超级个体"设计一个Logo。

[调用生成脚本]
python scripts/generate_image.py \
  --prompt "设计一个'超级个体'Logo，要求：
- 形状：圆形，简洁大方
- 颜色：蓝色和橙色渐变
- 元素：一个人的剪影 + AI元素（如电路板纹理）
- 风格：现代科技感，简约
- 寓意：一个人 + AI = 无限可能
- 用途：适合用在社交媒体头像、网站Logo" \
  --filename "super_individual_logo.png" \
  --resolution 2K

✅ 已生成Logo设计：super_individual_logo.png

Example 3: Social media illustration

User: "Generate a Xiaohongshu illustration, the theme is AI efficiency improvement"

Your reply：

好的，我来生成一张小红书配图。

[调用生成脚本]
python scripts/generate_image.py \
  --prompt "生成一张小红书配图，主题是'AI效率提升10倍'，要求：
- 尺寸：9:16竖版
- 配色：温暖的橙色和蓝色
- 文字：一个人 + AI = 效率提升10倍
- 元素：简约的人物剪影 + 上升的箭头
- 风格：现代简约，有科技感但不冰冷
- 适合小红书的视觉风格" \
  --filename "xiaohongshu_ai_efficiency.png" \
  --resolution 2K

✅ 已生成小红书配图：xiaohongshu_ai_efficiency.png

Environment configuration

Method 1: Use environment variables (recommended)

export NEXTAI_API_KEY="your-api-key-here"

Method 2: Command-line arguments

python scripts/generate_image.py \
  --prompt "你的提示词" \
  --api-key "your-api-key-here"

Method 3: Configuration file

In ~/.openclaw/openclaw.json configure:

{
  "skills": {
    "bananapro-image-gen": {
      "apiKey": "your-api-key-here",
      "baseUrl": "https://tryallapi.com",
      "model": "gemini-3-pro-image-preview"
    }
  }
}

Parameter description

Parameter	Description	Default value	Example
`--prompt`	Image description (required)	-	"Draw a cat"
`--filename`	Output filename	Auto-generated	"cat.png"
`--resolution`	Resolution	1K	1K/2K/4K
`--api-key`	API key	Environment variable	"sk-xxx"
`--model`	Model name	gemini-3-pro-image-preview	-

FAQ

Q: How can I improve generation quality?

A: Use detailed prompts and clearly specify:

Style (handwritten/modern/minimalist)
Color (blue/gradient/contrasting colors)
Elements (arrows/block diagrams/icons)
Composition (centered/left-right comparison/top-bottom layout)

Q: What should I do if generation fails?

A: Check:

Whether the API Key is correct
Whether the network connection is normal
Whether the prompt complies with the content policy

Q: Are Chinese prompts supported?

A: Fully supported! Gemini understands Chinese very well.

Q: How can I control costs?

Use 1K resolution for daily use (cheapest)
Use 2K/4K only when high definition is needed
Optimize prompts to reduce repeated generation

File structure

bananapro-image-gen/
├── SKILL.md              # 本文档（必需）
├── README.md             # 快速开始
├── scripts/
│   └── generate_image.py # 生成脚本
├── requirements.txt      # Python依赖
└── .gitignore           # Git忽略文件

Changelog

v1.0.0 (2026-02-11)

✅ Initial version
✅ Supports Gemini 3 Pro Image
✅ Supports multiple scenarios such as whiteboard images, logos, and illustrations
✅ Supports Chinese prompts
✅ Multiple resolution options

License

MIT License

Author

GitHub: @xianyu110
Tutorial: OpenClaw Super Individual Practical Guide

About this skill

Core Capabilities

Use Cases

Scenario 1: The user needs a tutorial illustration

Scenario 2: The user needs a logo design

Scenario 3: The user needs a comparison chart

How to Call

Basic call

High-definition images (2K/4K)

Configure the API using environment variables

Prompt Construction Techniques

Technique 1: Whiteboard diagram prompt template

Technique 2: Logo design prompt template

Technique 3: Comparison chart prompt template

Technique 4: Concept diagram prompt template

Workflow

Notes

Practical examples

Example 1: Tutorial illustration

Example 2: Logo design

Example 3: Social media illustration

Environment configuration

Method 1: Use environment variables (recommended)

Method 2: Command-line arguments

Method 3: Configuration file

Parameter description

FAQ

Q: How can I improve generation quality?

Q: What should I do if generation fails?

Q: Are Chinese prompts supported?

Q: How can I control costs?

File structure

Changelog

v1.0.0 (2026-02-11)

License

Author