← 返回 Skills 市场
163
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install zgcai-weekly-restaurant-menu-skill
功能描述
Extracts structured weekly menu content (breakfast, lunch, snacks by day) from a cafeteria menu image using a VLM via OpenRouter. Use when the user provides...
安全使用建议
This skill appears to do what it claims: it sends a user-supplied menu image to OpenRouter's VLM and returns structured Markdown. Before installing/running: 1) Be aware you are transmitting full images to an external service (privacy risk for sensitive images). 2) Confirm you trust the OpenRouter endpoint and restrict/rotate the API key if needed. 3) Fix the Python string quoting in scripts/extract_menu.py (the PROMPT contains unescaped/embedded quotes that will raise a syntax or runtime error) and test locally with non-sensitive images. 4) Monitor token usage/costs (the script reads usage/cost fields). If you need offline or private processing, request a version that runs locally or on a trusted private endpoint.
功能分析
Type: OpenClaw Skill
Name: zgcai-weekly-restaurant-menu-skill
Version: 1.0.0
The skill bundle is a straightforward implementation for extracting cafeteria menu data from images using the OpenRouter API. The Python script (scripts/extract_menu.py) uses standard libraries to send image data to a legitimate VLM endpoint (qwen/qwen-vl-plus) and contains no signs of data exfiltration, malicious execution, or prompt injection.
能力评估
Purpose & Capability
Name/description match the implementation: the script encodes an input image and calls OpenRouter's chat completions with a VLM model. The single required env var (MENU_OPENROUTER_API_KEY) is appropriate for calling OpenRouter.
Instruction Scope
SKILL.md and the script only perform the declared task (read image file, base64-encode, send to openrouter.ai, print structured Markdown). The script transmits the full image to an external service (OpenRouter) — a privacy/data-exfiltration consideration but expected for this capability. Additionally, the embedded PROMPT in scripts/extract_menu.py contains unescaped/internal quotes that will cause a Python syntax/runtime error as written; this is a correctness/quality issue that should be fixed before running.
Install Mechanism
No install spec (instruction-only + small helper script). Nothing is downloaded or written to disk by an installer; risk from install mechanism is minimal.
Credentials
Only MENU_OPENROUTER_API_KEY is required and that is the expected credential for invoking OpenRouter. No unrelated secrets or system paths are requested. Note: the API key will be sent as a Bearer token to openrouter.ai; treat it like any other credential (limit scope, rotate if exposed).
Persistence & Privilege
always is false and the skill does not request persistent/system-wide changes or modify other skills. It does not request elevated privileges or persistent presence.
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 zgcai-weekly-restaurant-menu-skill - 安装完成后，直接呼叫该 Skill 的名称或使用
/zgcai-weekly-restaurant-menu-skill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 release of zgcai-weekly-restaurant-menu-skill:
- Extracts structured weekly cafeteria menus from images (PNG/JPG), including breakfast, lunch, and snack stall items by day.
- Utilizes VLM via OpenRouter; requires a MENU_OPENROUTER_API_KEY for access.
- Outputs the week’s date range and dish lists in organized Markdown.
- Supports images up to 5 MB without compression.
- Snack stall items are listed separately and joined by "+" in output.
元数据
常见问题
中关村两院每周菜单识别 是什么？
Extracts structured weekly menu content (breakfast, lunch, snacks by day) from a cafeteria menu image using a VLM via OpenRouter. Use when the user provides... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 163 次。
如何安装 中关村两院每周菜单识别？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install zgcai-weekly-restaurant-menu-skill」即可一键安装，无需额外配置。
中关村两院每周菜单识别 是免费的吗？
是的，中关村两院每周菜单识别 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
中关村两院每周菜单识别 支持哪些平台？
中关村两院每周菜单识别 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 中关村两院每周菜单识别？
由 Jigao Fu（@ihainan）开发并维护，当前版本 v1.0.0。
推荐 Skills