Skillv1.0.1

ClawScan security

MenuVision · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

BenignFeb 23, 2026, 12:20 PM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: The skill's requirements and instructions are internally consistent: it needs python3 and a Google API key for Gemini Vision/Image calls and generates local scripts to build HTML menus; main user-facing risk is privacy of content sent to Google and any optional publishing step.
Guidance: This skill appears to do what it says, but consider these before installing: 1) Privacy: PDFs and photos you provide will be sent to Google's Gemini APIs for extraction and image generation — do not upload sensitive personal data unless you're comfortable with Google processing it. 2) API key scope: use a dedicated Google API key or service account with minimal required permissions and rotate it if shared. 3) Local scripts: the agent will generate and run Python scripts locally (ensure python3 is the expected interpreter). 4) Publishing: if you choose to publish to GitHub Pages, the agent should prompt for explicit GitHub credentials; do not provide tokens unless you trust the action. 5) Source provenance: the metadata points at a GitHub repo (github.com/ademczuk/MenuVision); if provenance matters, review that repository or contact the author before trusting the skill.

Review Dimensions

Purpose & Capability: okThe declared requirements (python3 and GOOGLE_API_KEY) match the SKILL.md: the pipeline extracts menu data via Gemini Vision, generates images via Gemini Image, and builds HTML using generated Python scripts. Requiring a Google API key is coherent for Gemini-based extraction and image generation, and python3 is reasonable because the agent will create and run Python scripts.
Instruction Scope: noteThe instructions confine themselves to extracting menu data from user-supplied URLs/PDFs/photos, generating images, building local HTML, and optionally publishing to GitHub Pages. However, the extraction and image-generation steps explicitly send provided menu content (images/PDF text) to Google's Gemini APIs — users should be aware their menu content (and any embedded data in PDFs/photos) will be transmitted to Google. The SKILL.md does not instruct reading unrelated local files or environment variables.
Install Mechanism: okThis is an instruction-only skill with no install spec and no code files, so nothing is downloaded or written by the installer itself. That keeps install risk low; the agent will generate Python scripts at runtime which will run locally under the user's environment.
Credentials: okOnly one required env var (GOOGLE_API_KEY) is declared and used for Gemini calls — proportionate to the stated functionality. No unrelated credentials (e.g., AWS, GitHub) are required. Note: an optional publish step to GitHub Pages is listed but no GitHub token is requested by default; if the agent attempts to publish, it should request explicit permission and the appropriate GitHub credentials at that time.
Persistence & Privilege: okThe skill does not request permanent/always-enabled presence. It does not modify other skills or global agent settings per the SKILL.md. Runtime-generated scripts operate within the user's environment when invoked.