xlsx-skill

Security checks across static analysis, malware telemetry, and agentic risk

Overview

The skill is a normal Excel-spreadsheet helper, with expected cautions about installing dependencies and allowing local spreadsheets to be read or written.

This skill appears safe for normal spreadsheet work. Before installing, verify the Python and optional LibreOffice dependencies, use a virtual environment if possible, and only let the agent read or edit spreadsheets whose contents you are comfortable sharing in the session.

Static analysis

No static analysis findings were reported for this release.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Risk analysis

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

#
ASI02: Tool Misuse and Exploitation
Low
What this means

If used on private or important spreadsheets, the agent may expose their contents in conversation or create/overwrite output files.

Why it was flagged

The skill teaches the agent how to read and save spreadsheet files. This is central to the skill's purpose, but it means user-invoked use can access or modify local files.

Skill content
df = pd.read_excel('file.xlsx') ... wb.save('output.xlsx')
Recommendation

Use explicit filenames and output paths, work on copies of important spreadsheets, and review generated files before relying on them.

#
ASI04: Agentic Supply Chain Vulnerabilities
Low
What this means

Installing these dependencies can change the local Python or Homebrew environment and relies on external package sources.

Why it was flagged

The skill asks the user to install third-party packages and an optional external application. This is expected for spreadsheet processing, but versions are not pinned and installation is outside a formal install spec.

Skill content
pip install pandas openpyxl xlsxwriter ... brew install --cask libreoffice
Recommendation

Install dependencies from trusted package managers, preferably in a virtual environment, and pin versions if reproducibility matters.