ClawHub
Back to skill

Security audit

Excel Data Assistant / Excel 数据处理

Security checks across malware telemetry and agentic risk

Overview

This skill is mainly an upsell/advertising stub that claims Excel automation but does not provide working Excel-processing instructions or tools.

Review before installing. This does not appear to contain malware or privileged behavior, but it should not be relied on to process Excel files as advertised; expect promotional links and a paid-version upsell rather than functional spreadsheet automation.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Description-Behavior Mismatch

Medium
Confidence
98% confidence
Finding
The skill advertises Excel processing capabilities, but the documented commands only print status text, promotional copy, and external links. This is a deceptive or materially misleading implementation that can cause users or downstream agents to rely on functionality that does not exist, potentially triggering incorrect automation decisions or sending users to third-party services instead of performing the requested task.

Intent-Code Divergence

Medium
Confidence
97% confidence
Finding
The documentation presents the skill as operational and ready for Excel work, but the usage section only advertises a paid/full version and does not implement the claimed data-processing features. In an agent ecosystem, this can misdirect workflows, waste execution time, and route users toward external commercial endpoints under false pretenses.

Vague Triggers

Medium
Confidence
80% confidence
Finding
The description uses broad language implying general Excel processing without clear scope, triggers, or operational boundaries. While not directly exploitable on its own, this increases the chance of over-invocation and user or agent misinterpretation, especially given the mismatch between advertised and actual behavior.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.