Skillv2.0.0

ClawScan security

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

Scanner verdict

BenignMar 10, 2026, 2:23 AM

Verdict: benign
Confidence: medium
Model: gpt-5-mini
Summary: The skill appears to implement exactly what it claims — a Python-based Excel parsing utility — with no obvious attempts to access unrelated secrets or external endpoints, but it will auto-install Python packages at runtime which has operational risk and requires internet access.
Guidance: This skill looks consistent with its stated purpose, but be aware it will attempt to pip-install missing dependencies at runtime (internet access to PyPI is required). To reduce risk: run it in a sandboxed virtual environment, pin package versions, inspect the required packages (python-calamine, xlrd, openpyxl) before allowing installs, and verify the full script (including the truncated portion) if you need higher assurance. Also note the repository/homepage is unknown and setup.py lists version 1.0.0 while the registry metadata shows 2.0.0 — prefer obtaining packages from a trusted source (official PyPI/GitHub) before use.

Review Dimensions

Purpose & Capability: okName, description, SKILL.md, requirements.txt and the Python code all align: this is an Excel parsing skill that uses python-calamine with xlrd/openpyxl as fallbacks. The code and declared dependencies are proportionate to the stated purpose.
Instruction Scope: noteSKILL.md and the code only describe reading user-supplied Excel files and optional .env settings (EXCEL_MAX_ROWS, EXCEL_KEEP_EMPTY_ROWS). The runtime code parses only the provided file paths and does not reference unrelated system paths or external endpoints. Note: the code calls load_dotenv() and will read .env if present (expected behavior).
Install Mechanism: noteThere is no separate install spec, but the shipped Python script contains an install_dependency() helper that runs pip install via subprocess at runtime to fetch missing packages (e.g., python-calamine, xlrd). Using pip at runtime is a common convenience but increases risk because it performs network fetches and executes third-party package code; this is moderate risk but coherent with the skill's need for optional dependencies.
Credentials: okThe skill does not declare required environment variables or credentials in the registry. It documents optional .env keys for parser behavior (max rows, keep empty rows), which are reasonable and limited. No secrets, API keys, or unrelated env variables are requested.
Persistence & Privilege: okalways is false and the skill does not request elevated platform privileges. The only persistence-like behavior is installing Python packages into the runtime environment when dependencies are missing, which is expected but changes the environment; it does not modify other skills or system-wide agent settings.