← 返回 Skills 市场
299
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install nano-banana-2-fal
功能描述
Generate or edit images via Gemini 3 Pro Image (Nano Banana Pro).
安全使用建议
This skill is internally consistent: it legitimately needs GEMINI_API_KEY and the 'uv' runner to execute the bundled Python script which calls Google's genai library. Before installing, verify the source of the 'uv' Homebrew formula on your machine (so you know what will be installed), ensure your Python environment will provide google-genai and pillow (or that 'uv run' will handle deps), and supply a Gemini API key with appropriate, minimal permissions (rotate or limit the key if possible). Also be mindful that the script will read any input image files you pass to it — do not feed sensitive images you don't want transmitted to the Gemini API. If you want higher assurance, run the script locally after review rather than granting broad automated access to the skill.
功能分析
Type: OpenClaw Skill
Name: nano-banana-2-fal
Version: 1.0.0
The skill bundle provides a straightforward implementation for generating and editing images using the Google GenAI SDK. The Python script `scripts/generate_image.py` correctly handles API authentication, image processing via Pillow, and interaction with the Gemini API without any signs of data exfiltration, obfuscation, or unauthorized execution. While the model name 'gemini-3-pro-image-preview' and the 'Nano Banana' branding appear to be placeholders or fictional, the code logic is transparent and aligns with the stated purpose in `SKILL.md`.
能力评估
Purpose & Capability
Name/description match the bundled Python script which calls the Google genai client (model 'gemini-3-pro-image-preview') and accepts image input(s) and a prompt. The required binary 'uv' is used in SKILL.md to execute the script, and GEMINI_API_KEY is the only credential needed — all of which are coherent with an image-generation skill.
Instruction Scope
SKILL.md only instructs running the included script via 'uv' and to provide a prompt, filename, optional input images, and resolution. The script reads only provided input image file paths and the GEMINI API key (or an explicit --api-key), uploads/requests generation via the GenAI client, saves the returned image locally, and prints a MEDIA token. There are no instructions to read unrelated files, harvest other environment variables, or send data to unexpected endpoints.
Install Mechanism
Install spec uses brew to install a formula named 'uv' and declares it will create a 'uv' binary. Using Homebrew is a normal distribution path; this is not an arbitrary URL download. If you are unfamiliar with the 'uv' formula on your platform, confirm the formula source/tap before installing. The Python dependencies (google-genai, pillow) are declared in the script header but not managed by the brew install — ensure your environment or uv run will install/resolve them.
Credentials
Only GEMINI_API_KEY is required (and an optional --api-key argument). That single credential is appropriate and necessary for calling Google's Gemini/GenAI API. The script does not request other unrelated secrets or config paths.
Persistence & Privilege
The skill is not always-enabled and uses the platform default (agent can invoke it autonomously). It does not modify other skills or system-wide configuration and does not request persistent privileges beyond its own runtime.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install nano-banana-2-fal - 安装完成后，直接呼叫该 Skill 的名称或使用
/nano-banana-2-fal触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.0
Initial publish
元数据
常见问题
Nano Banana 2 FAL 是什么？
Generate or edit images via Gemini 3 Pro Image (Nano Banana Pro). 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 299 次。
如何安装 Nano Banana 2 FAL？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install nano-banana-2-fal」即可一键安装，无需额外配置。
Nano Banana 2 FAL 是免费的吗？
是的，Nano Banana 2 FAL 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Nano Banana 2 FAL 支持哪些平台？
Nano Banana 2 FAL 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Nano Banana 2 FAL？
由 wingchiu（@wingchiu）开发并维护，当前版本 v1.0.0。
推荐 Skills