ClawHub
Back to skill

Security audit

docx-trackdiff

Security checks across malware telemetry and agentic risk

Overview

This skill locally compares two user-provided DOCX files and writes a tracked-changes DOCX, with no evidence of hidden data access, network upload, persistence, or unrelated behavior.

Install only if you want an agent to process local DOCX drafts for redline comparison. Because DOCX files can contain sensitive text and embedded media, explicitly name the two intended files and review the generated tracked-changes document in Word for high-stakes work.

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 (1)

Vague Triggers

Medium
Confidence
87% confidence
Finding
The README advertises very broad trigger phrases such as comparing 'these two Word documents' and says the agent will automatically read SKILL.md, invoke scripts, and return an output file. Without tighter scope constraints and explicit input-validation guidance, this can cause over-activation or activation on untrusted/mis-scoped files, increasing the chance the skill processes sensitive documents unexpectedly or is invoked in workflows where safer review should occur first.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.