← 返回 Skills 市场
110
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install openclaw-pro-nano-banana-pro-1-0-2
功能描述
Generate or edit images via Gemini 3 Pro Image (Nano Banana Pro).
安全使用建议
This skill mostly looks like a legitimate Gemini image helper, but there are a few packaging inconsistencies you should check before installing: 1) Confirm you trust the source—registry ownerId vs _meta.json ownerId differ. 2) Ensure you are comfortable providing GEMINI_API_KEY (this is the only credential requested). 3) The SKILL.md and script disagree about Python dependencies: the script needs google-genai and pillow, while SKILL.md lists requests and beautifulsoup4; your environment must have the actual Python packages installed or the script will fail. 4) The install step only installs the 'uv' brew formula; it will not install Python packages automatically. If those points are acceptable (or you can fix the dependencies locally), the skill is reasonable to use. If you need higher assurance, ask the author for a corrected package manifest and consistent metadata (matching ownerId and explicit Python install steps).
功能分析
Type: OpenClaw Skill
Name: openclaw-pro-nano-banana-pro-1-0-2
Version: 1.0.2
The skill bundle is a standard wrapper for an image generation API (Gemini). The Python script `scripts/generate_image.py` uses the official `google-genai` and `pillow` libraries to process prompts and images, saving the results locally and emitting a standard `MEDIA:` tag for the OpenClaw agent. While `SKILL.md` lists unused dependencies (`requests`, `beautifulsoup4`) and the metadata contains a future-dated timestamp, there is no evidence of data exfiltration, malicious execution, or prompt injection.
能力评估
Purpose & Capability
Name/description, required binary (uv), and primary env var (GEMINI_API_KEY) align with the included script which calls google.genai. However the SKILL.md metadata lists dependencies (requests, beautifulsoup4) that the script never imports, while the script header and imports require google-genai and pillow. Also the ownerId in _meta.json differs from the registry ownerId in the provided metadata—these mismatches are packaging/authoring inconsistencies.
Instruction Scope
Runtime instructions are narrowly scoped: run the bundled script with prompt/filename and optional input images. The script only reads input images you explicitly pass, uses the GEMINI_API_KEY or --api-key, uploads content to Google's GenAI client, and writes the generated PNG to the requested output path. It prints a MEDIA:<path> token for attachment and does not attempt to read unrelated files or exfiltrate other environment variables.
Install Mechanism
Install spec only installs the 'uv' brew formula, which matches the expected runtime invocation. There is no automated installation declared for the Python packages the script needs (google-genai, pillow). That mismatch means the script may fail at runtime unless the environment already provides those Python deps; it's an operational inconsistency rather than an explicit security issue.
Credentials
Only GEMINI_API_KEY is required (declared as primaryEnv). The script uses only that key (or --api-key). No other credentials, secrets, or unrelated environment variables are requested.
Persistence & Privilege
The skill does not request always:true, does not modify other skills or global configs, and does not persist credentials itself. It only writes generated image files to the user-specified output path.
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 openclaw-pro-nano-banana-pro-1-0-2 - 安装完成后，直接呼叫该 Skill 的名称或使用
/openclaw-pro-nano-banana-pro-1-0-2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
v1.0.2
- Added detailed usage instructions for Nano Banana Pro, including image generation, editing, and multi-image composition.
- Documented required dependencies, runner, and API key setup (GEMINI_API_KEY).
- Outlined supported resolutions and aspect ratios, with flags for custom settings.
- Provided guidance for filename conventions and output handling.
- Included installation steps for required tools (uv via brew).
元数据
常见问题
OpenClaw Pro 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 110 次。
如何安装 OpenClaw Pro？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install openclaw-pro-nano-banana-pro-1-0-2」即可一键安装，无需额外配置。
OpenClaw Pro 是免费的吗？
是的，OpenClaw Pro 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
OpenClaw Pro 支持哪些平台？
OpenClaw Pro 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 OpenClaw Pro？
由 jeayoo（@jeayoo）开发并维护，当前版本 v1.0.2。
推荐 Skills