← 返回 Skills 市场
guyoung

OpenCV

作者 guyoung · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ⚠ suspicious
118
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install opencv
功能描述
Computer vision and image processing using OpenCV WebAssembly. Uses opencv-component.wasm running in openclaw-wasm-sandbox plugin. Supports image processing,...
安全使用建议
This skill appears to do what it says (run OpenCV in a WASM sandbox), but before installing or using it you should: 1) verify the WASM binary's provenance — prefer official releases or a checksum/signature; 2) avoid mapping wide host paths (don't expose your entire home or system directories); map only the specific image directories needed; 3) confirm the openclaw-wasm-sandbox enforces isolation (no network access and restricted file access) or explicitly limit allowed-outbound-hosts before running model/dnn operations; 4) if possible, inspect the WASM source or run it in a disposable environment first; and 5) prefer a skill that provides an install spec with signed artifacts or an official homepage/repository. These precautions reduce the risk of executing untrusted code or accidental data exposure.
功能分析
Type: OpenClaw Skill Name: opencv Version: 1.0.0 The skill provides a legitimate interface for computer vision tasks using OpenCV compiled to WebAssembly. It utilizes the 'openclaw-wasm-sandbox' to execute operations like image processing, object detection, and machine learning in a restricted environment. No evidence of malicious intent, data exfiltration, or prompt injection was found; the instructions and documentation (SKILL.md and references/operations.md) are strictly aligned with the stated purpose of providing image manipulation tools.
能力评估
Purpose & Capability
Name and description match the instructions: the skill expects an OpenCV WebAssembly component and the openclaw-wasm-sandbox plugin. Declared prerequisites (plugin + wasm file) are consistent with the stated functionality.
Instruction Scope
Runtime instructions are narrowly scoped to invoking the wasm component via wasm-sandbox-run and mapping workDir/mapDir for file I/O. The skill does instruct mapping arbitrary host directories into the sandbox (expected for image files) — this is normal for the task but increases the risk of exposing unrelated sensitive files if mappings are too broad.
Install Mechanism
No formal install spec is present; SKILL.md instructs downloading a WASM file from a raw.githubusercontent.com URL. While GitHub raw is a common host, the skill provides no checksum/signature or provenance guidance. Downloading and executing an externally-hosted WASM blob is moderate-to-high risk if unverified.
Credentials
The skill requests no environment variables, credentials, or config paths. This is proportionate to an image-processing WASM component. Note: the ability to map host directories into the sandbox effectively grants the component access to any mapped files.
Persistence & Privilege
No always:true flag, no install-time modifications described, and the skill is instruction-only. It does not request persistent elevated privileges via the metadata.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install opencv
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /opencv 触发
  4. 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
OpenCV computer vision skill initial release: - Brings comprehensive image processing and computer vision capabilities via OpenCV WebAssembly. - Supports operations such as object detection, feature extraction, panorama stitching, photo enhancement, camera calibration, and machine learning. - Enables batch processing with multi-threading support. - Integrates through the openclaw-wasm-sandbox plugin using the opencv-component.wasm file. - Includes detailed usage instructions and a categorized operations overview.
元数据
Slug opencv
版本 1.0.0
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 1
常见问题

OpenCV 是什么?

Computer vision and image processing using OpenCV WebAssembly. Uses opencv-component.wasm running in openclaw-wasm-sandbox plugin. Supports image processing,... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 118 次。

如何安装 OpenCV?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install opencv」即可一键安装,无需额外配置。

OpenCV 是免费的吗?

是的,OpenCV 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

OpenCV 支持哪些平台?

OpenCV 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 OpenCV?

由 guyoung(@guyoung)开发并维护,当前版本 v1.0.0。

💬 留言讨论