Security audit

Excel大师

Security checks across malware telemetry and agentic risk

Overview

This Excel automation skill is purpose-aligned and disclosed, with ordinary spreadsheet file-reading and Python execution guidance but no hidden persistence or exfiltration behavior.

Before installing, treat it as a local spreadsheet automation helper: use explicit input and output paths, avoid overwriting original workbooks unless you have backups, and review any generated Python command before running it on sensitive or bulk files.

SkillSpector

By NVIDIA

Vulnerability Patterns

Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
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 (2)

Vague Triggers

Medium

Confidence: 88% confidence
Finding: The trigger list is very broad and consists of common spreadsheet terms such as Excel, xlsx, 合并, 拆分, and pandas. In an agent ecosystem, this can cause the skill to activate in many routine conversations, increasing the chance that powerful file-reading or code-execution behavior is invoked when the user did not specifically intend to use this skill.

Missing User Warnings

Medium

Confidence: 91% confidence
Finding: The skill explicitly discusses overwriting existing files and creating batch outputs, but does not consistently require a user-facing confirmation, dry-run mode, backup strategy, or safe output isolation. In a skill with exec capability and filesystem access, that omission can lead to accidental data loss or unintended modification of many files if the agent acts on ambiguous instructions.

VirusTotal

59/59 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.