使用Nano Banana Pro（Gemini 3 Pro Image）通过第三方中转API站清云EchoFlow API生成或编辑图像。支持图像生成、图像编辑和多图像合成（最多14张图像）。当用户请求生成图像、创建图像、编辑图像或组合图像时使用。关键词：图像生成、图片生成、生成图片、AI绘画、Nano Banana Pro、Gemini、Gemini 3 Pro Image。

通过清云 EchoFlow API 生成和编辑图片（基于 Nano Banana Pro / Gemini 3 Pro Image）。 支持图片生成、单图编辑、多图合成（最多14张）、多种分辨率（1K/2K/4K）。 触发词：图像生成、图片生成、AI绘画、生成美女、生成图片、Nano Banana Pro、Gem...

Security Scan

The skill is internally consistent: it implements image generation/editing via EchoFlow, only requires the EchoFlow API key and a uv runtime, and its instructions and code match the described purpose.

Purpose & Capability

Name/description match the included code and instructions: the script calls EchoFlow endpoints to generate/edit images and supports multi-image composition and resolution options. Required items (uv, ECHOFLOW_API_KEY) are appropriate for this functionality.

Instruction Scope

SKILL.md and the Python script stay within scope: they read provided input image files, use only ECHOFLOW_API_KEY (or --api-key), and post to the configured API base. Instructions explicitly warn about API base and CLI key exposure. The skill does not access unrelated system files or other environment variables.

Install Mechanism

There is no bundled install spec beyond recommending installing the 'uv' runtime (pip/brew/scoop or a curl | sh installer). Recommending running a remote install script via curl|sh is a legitimate convenience but carries standard supply-chain risk—users should inspect the installer source before running it.

Credentials

Only ECHOFLOW_API_KEY is required and declared as the primary credential. The code explicitly avoids falling back to unrelated keys (OPENAI_API_KEY/GEMINI_API_KEY). This credential request is proportional to the stated purpose.

Persistence & Privilege

The skill is not always-on and does not request elevated or persistent system privileges, nor does it modify other skills' configs or request filesystem paths beyond reading user-specified images and writing output files.

Assessment

This skill appears to do what it claims, but before installing: 1) only provide a trusted EchoFlow API key and prefer setting it via the ECHOFLOW_API_KEY environment variable (avoid --api-key to prevent exposure in process lists); 2) keep --api-base as the default (https://api.echoflow.cn/v1) — do not point it at unknown hosts because your key will be sent there; 3) inspect the 'uv' installer script if you plan to run curl | sh; and 4) be aware that images and any input files you supply will be uploaded to the EchoFlow service per the script behavior and EchoFlow's policies.

EchoFlow 图片生成 (Nano Banana Pro) / Image Generation

通过清云 EchoFlow API 使用 Nano Banana Pro (Gemini 3 Pro Image) 生成或编辑图片。

Generate or edit images using EchoFlow API with Nano Banana Pro (Gemini 3 Pro Image).

需求 / Requirements

• uv 运行时 — 用于执行捆绑的 Python 脚本，自动管理依赖 uv runtime — runs the bundled Python script with auto-managed dependencies
• ECHOFLOW_API_KEY — 你的清云 API 密钥 / Your EchoFlow API key: https://api.echoflow.cn/
• Python 包（由 uv 自动安装）: httpx>=0.25.0, pillow>=10.0.0 (auto-installed by uv)

安装 / Setup

第一步 — 安装 uv / Install uv:

# Windows (pip)
pip install uv

# Windows (Scoop)
scoop install uv

# macOS (Homebrew)
brew install uv

# Linux (curl | sh — 检查脚本后再运行 / Inspect script before running)
curl -LsSf https://astral.sh/uv/install.sh | sh

第二步 — 设置 API 密钥 / Set your API key:

# Windows — 当前会话 / Current session
$env:ECHOFLOW_API_KEY = "sk-..."

# Windows — 永久设置 / Permanent (user-level)
[Environment]::SetEnvironmentVariable("ECHOFLOW_API_KEY", "sk-...", "User")

# macOS / Linux
export ECHOFLOW_API_KEY="sk-..."

⚠️ 安全建议 / Security note: 优先使用环境变量而非 --api-key，因为命令行参数会暴露在进程列表和 shell 历史中。
Prefer the ECHOFLOW_API_KEY env var over --api-key to avoid key exposure in process lists and shell history.

快速开始 / Quick Start

生成图片 / Generate Image

uv run {baseDir}/scripts/generate_image.py --prompt "a serene mountain landscape at sunset" --filename "mountain.png"

编辑单张图片 / Edit Single Image

uv run {baseDir}/scripts/generate_image.py --prompt "add a rainbow in the sky" --filename "edited.png" -i "input.png"

多图合成（最多14张）/ Multi-Image Composition (up to 14)

uv run {baseDir}/scripts/generate_image.py --prompt "combine these into one scene" --filename "combined.png" -i img1.png -i img2.png -i img3.png

参数 / Parameters

参数	描述 / Description	默认值
--prompt, -p	图片描述（必填）	—
--filename, -f	输出文件名（必填）	—
--input-image, -i	输入图片用于编辑/合成（可重复，最多14张）	—
--resolution, -r	输出分辨率: 1K, 2K, 4K	1K
--model, -m	模型名称	gemini-3.1-flash-image-preview
--api-key, -k	API 密钥（优先使用环境变量）	—
--api-base	API 地址（见安全警告）	https://api.echoflow.cn/v1

可用模型 / Available Models

• gemini-3.1-flash-image-preview (默认) — 更快，更稳定 / faster, more stable
• gemini-3-pro-image-preview — 更高质量，可能遇到 429 限流 / higher quality, may hit 429

分辨率 / Resolutions

• 1K (默认 / default) — 标准画质
• 2K — 高画质 / high quality
• 4K — 超高画质 / ultra high quality

自动检测 / Auto-detection: 编辑图片时，分辨率根据最大输入图片尺寸自动调整（≥3000px → 4K, ≥1500px → 2K, 否则 1K）

安全说明 / Security Notes

API 密钥范围: 脚本仅从 ECHOFLOW_API_KEY 环境变量或 --api-key 参数读取密钥，不会回退到 OPENAI_API_KEY 或 GEMINI_API_KEY。

--api-base 风险: API 密钥会以 Authorization 头发送到指定的主机。不要将其指向不信任的端点，始终使用默认地址 https://api.echoflow.cn/v1 或你完全信任的主机。

CLI 密钥暴露: 通过 --api-key 传入密钥会暴露在进程列表和 shell 历史中。请使用 ECHOFLOW_API_KEY 环境变量。

示例 / Examples

# 简单生成
uv run {baseDir}/scripts/generate_image.py -p "a cute cat wearing a hat" -f "cat.png"

# 高分辨率
uv run {baseDir}/scripts/generate_image.py -p "futuristic city" -f "city.png" -r 4K

# 编辑单图 — 添加雪花
uv run {baseDir}/scripts/generate_image.py -p "add snow to the scene" -f "snowy.png" -i summer.png

# 多图合成
uv run {baseDir}/scripts/generate_image.py -p "create a collage" -f "collage.png" -i a.png -i b.png -i c.png

# 使用 Pro 模型
uv run {baseDir}/scripts/generate_image.py -p "abstract art" -f "art.png" -m "gemini-3-pro-image-preview"

API 参考 / API Reference

详细 API 文档见 / See detailed API docs at: echoflow_api.md