Security audit

信创文档编辑 Skill

Security checks across malware telemetry and agentic risk

Overview

This is a small Word document editing skill with no executable code; the main caveat is that its instructions are sparse and incomplete.

Safe to install from a security perspective, but keep backups before editing important Word documents and review the output carefully because the skill’s replacement guidance is incomplete.

SkillSpector

By NVIDIA

Vulnerability Patterns

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
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Vague Triggers

Medium

Confidence: 91% confidence
Finding: The skill description is underspecified and broadly scoped to editing Word documents without clear activation boundaries, allowed actions, or safety constraints. Overly broad skills can be invoked in unintended contexts, increasing the chance of unauthorized document modification, misuse on sensitive files, or ambiguous agent behavior.

Natural-Language Policy Violations

Medium

Confidence: 83% confidence
Finding: The description states the skill is specifically for Word document editing in Chinese without indicating user choice or a justified localization requirement. Forcing a language can cause user confusion, misinterpretation of edits, and unsafe operation if the agent transforms or presents document content in a language the user did not request.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.