ClawHub
Back to skill

Security audit

Excel AI Analyzer

Security checks across malware telemetry and agentic risk

Overview

This skill reads user-selected Excel files to produce local statistical summaries, with no evidence of hidden network access, credentials, persistence, or destructive behavior.

Install only if you are comfortable letting the agent read the Excel files you provide. Avoid using it on spreadsheets containing secrets, regulated personal data, or confidential financial/business information unless your agent environment is approved for that data; also do not rely on the advertised chart-generation feature in this version.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Missing User Warnings

Medium
Confidence
84% confidence
Finding
This skill processes Excel files, which commonly contain sensitive business, financial, or personal data, yet the README provides no warnings about privacy, confidentiality, or careful handling of uploaded content. In practice, users may submit regulated or confidential spreadsheets without understanding the risks, increasing the likelihood of data exposure through logs, reports, or downstream tooling.

Vague Triggers

Medium
Confidence
87% confidence
Finding
The invocation example uses a very broad natural-language trigger ('帮我分析这个 Excel 文件') without any activation boundaries, confirmation requirements, or scoping rules. In agent environments, overly generic triggers can cause unintended skill activation on ordinary user requests, increasing the chance of processing sensitive local files or performing analysis when the user did not explicitly intend to invoke this skill.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.