← Back to Skills Marketplace
xianyu110

Bananapro Image Gen

by xianyu110 · GitHub ↗ · v1.0.2 · MIT-0
cross-platform ⚠ suspicious
282
Downloads
0
Stars
0
Active Installs
3
Versions
Install in OpenClaw
/install bananapro-image-gen
Description
使用 Gemini 图像模型生成图片,支持白板图、Logo设计、社交媒体配图。
README (SKILL.md)

Banana Pro Image Generation Skill

使用 Gemini 图像模型生成各种风格的图片。

功能特点

  • 文生图:根据文字描述生成图片
  • 白板图:生成手写风格的概念图、流程图
  • Logo设计:创建简约现代的Logo
  • 社交媒体配图:生成适合各平台的配图
  • 图片编辑:修改已有图片

中转 API 配置教程

什么是中转 API?

中转 API 是第三方提供的 API 代理服务,相比官方 API 有以下优势:

  • 成本更低(通常 5-7 折)
  • 国内网络直连,更稳定
  • 无需配置代理

推荐中转服务

apipro.maynor1024.live(默认)

网站:https://apipro.maynor1024.live/

配置方式

方式一:使用中转 API(推荐)

export NEXTAI_API_KEY="your-api-key"

获取 API Key:

  1. 访问 https://apipro.maynor1024.live/
  2. 注册账号
  3. 获取 API Key

方式二:使用官方 Gemini API

export GEMINI_API_KEY="your-gemini-api-key"

获取方式:访问 https://aistudio.google.com/apikey

注意:官方 API 需要科学上网。

方式三:使用其他中转服务

支持任何 OpenAI 兼容的中转服务:

export NEXTAI_API_KEY="your-api-key"
export NEXTAI_API_URL="https://your-proxy.com/v1"

使用方法

基础用法

python scripts/generate_image.py \
  --prompt "A serene Japanese garden" \
  --filename "garden.png"

高清图片

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

图片编辑

python scripts/generate_image.py \
  --prompt "把天空改成夕阳效果" \
  --filename "edited.png" \
  --input-image "original.jpg"

分辨率选项

  • 1K(默认):~1024px,日常使用
  • 2K:~2048px,高清展示
  • 4K:~4096px,打印输出

提示词技巧

白板图模板

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]
- 使用箭头、框图等手绘元素

Logo设计模板

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/简约]

常见问题

Q: 中转 API 安全吗?

A: 选择可信的中转服务很重要。apipro.maynor1024.live 是社区常用的中转服务。

Q: 支持中文提示词吗?

A: 完全支持!Gemini 对中文理解很好。

许可证

MIT License

Usage Guidance
This skill appears to implement Gemini-style image generation and will work if you provide an API key and requests. However: 1) the package metadata failing to declare the required API key is a red flag — assume you must supply NEXTAI_API_KEY or GEMINI_API_KEY to run it; 2) SKILL.md recommends and defaults to a third‑party proxy (apipro.maynor1024.live). If you use that proxy or any custom NEXTAI_API_URL, you will send your API key, prompts, and any uploaded images to that provider — only use a proxy you fully trust. 3) If you want better safety, use the official GEMINI_API_KEY and the official Google generativelanguage endpoint (or your own trusted proxy), and avoid submitting private/sensitive images or prompts to unknown services. 4) There is a minor mismatch in the script's payload/response field names that may cause failures; consider reviewing or testing the script before running at scale.
Capability Analysis
Type: OpenClaw Skill Name: bananapro-image-gen Version: 1.0.2 The skill provides image generation via Gemini but strongly encourages users to route their API keys through a third-party proxy service (apipro.maynor1024.live). The script `scripts/generate_image.py` transmits the API key as a plaintext query parameter to the configured endpoint, which is a security risk. While the code performs its stated function, the promotion of an unofficial intermediary to handle sensitive credentials creates a high potential for credential harvesting.
Capability Assessment
Purpose & Capability
Name/description match the included script: it generates images with a Gemini-style API. However the registry metadata claims no required env vars or primary credential while the code requires an API key (NEXTAI_API_KEY or GEMINI_API_KEY). This mismatch is inconsistent and should be corrected.
Instruction Scope
SKILL.md instructs users to register at and use a third‑party proxy (apipro.maynor1024.live) by setting NEXTAI_API_KEY/NEXTAI_API_URL. The script will POST prompts and (optionally) base64-encoded input images to whatever NEXTAI_API_URL is set to, so using a proxy hands your API key and content to that provider. The instructions also default to that proxy as 'recommended' which increases risk.
Install Mechanism
No install spec; only a small Python script and requests dependency. Risk from installation is low (user only needs to pip install requests).
Credentials
The script requires NEXTAI_API_KEY or GEMINI_API_KEY and optionally NEXTAI_API_URL, but the skill metadata lists no required env vars or primary credential. Requiring API keys is reasonable for this purpose, but the missing metadata and the encouragement to use an external proxy that will receive those keys is disproportionate and risky.
Persistence & Privilege
Skill is not always-enabled and does not request elevated or persistent platform privileges. It does not modify other skills or system settings.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install bananapro-image-gen
  3. After installation, invoke the skill by name or use /bananapro-image-gen
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.2
完善中转 API 配置教程,添加 apipro.maynor1024.live
v1.0.1
添加中转 API 配置教程和常见问题解答
v1.0.0
首个版本:支持 Gemini 图像生成、白板图、Logo设计、社交媒体配图
Metadata
Slug bananapro-image-gen
Version 1.0.2
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 3
Frequently Asked Questions

What is Bananapro Image Gen?

使用 Gemini 图像模型生成图片,支持白板图、Logo设计、社交媒体配图。 It is an AI Agent Skill for Claude Code / OpenClaw, with 282 downloads so far.

How do I install Bananapro Image Gen?

Run "/install bananapro-image-gen" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Bananapro Image Gen free?

Yes, Bananapro Image Gen is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Bananapro Image Gen support?

Bananapro Image Gen is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Bananapro Image Gen?

It is built and maintained by xianyu110 (@xianyu110); the current version is v1.0.2.

💬 Comments