← Back to Skills Marketplace
guyoung

OpenCV

by guyoung · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ⚠ suspicious
118
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install opencv
Description
Computer vision and image processing using OpenCV WebAssembly. Uses opencv-component.wasm running in openclaw-wasm-sandbox plugin. Supports image processing,...
Usage Guidance
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.
Capability Analysis
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.
Capability Assessment
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.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install opencv
  3. After installation, invoke the skill by name or use /opencv
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
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.
Metadata
Slug opencv
Version 1.0.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is OpenCV?

Computer vision and image processing using OpenCV WebAssembly. Uses opencv-component.wasm running in openclaw-wasm-sandbox plugin. Supports image processing,... It is an AI Agent Skill for Claude Code / OpenClaw, with 118 downloads so far.

How do I install OpenCV?

Run "/install opencv" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is OpenCV free?

Yes, OpenCV is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does OpenCV support?

OpenCV is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created OpenCV?

It is built and maintained by guyoung (@guyoung); the current version is v1.0.0.

💬 Comments