Vision Tagger

Tag and annotate images using Apple Vision framework (macOS only). Detects faces, bodies, hands, text (OCR), barcodes, objects, scene labels, and saliency re...

This skill appears to do local image analysis using Apple Vision and Pillow; it requires macOS 12+ and Xcode CLI tools. Before installing: (1) only run on macOS as intended (SKILL.md requires macOS); (2) review the included Swift and Python source (they are provided) and only run the compile/install steps if you trust the source; (3) be aware the setup compiles a binary in the skill folder and installs Pillow via pip; (4) run the tool on non-sensitive images first to confirm behavior; and (5) if you want extra caution, run the setup/annotation inside a sandboxed account or VM and inspect the compiled binary with standard tools.

Type: OpenClaw Skill Name: vision-tagger Version: 1.0.0 The OpenClaw skill 'vision-tagger' is benign. All files (SKILL.md, setup.sh, annotate_image.py, image_tagger.swift) align with the stated purpose of local image analysis using Apple's Vision framework. The `SKILL.md` and `setup.sh` contain standard commands for installing Xcode CLI tools and Python's Pillow library, and for compiling the Swift binary. The Python script `annotate_image.py` uses `subprocess.run` to execute the Swift binary, but passes arguments as a list, mitigating shell injection risks. The core Swift binary `image_tagger.swift` uses only Apple's Vision and AppKit frameworks for image processing, without any network calls, access to sensitive user data, persistence mechanisms, or obfuscation. There is no evidence of prompt injection or intentional harmful behavior.