ClawHub
Back to skill

Security audit

敏感词筛查 CMT 1.0

Security checks across malware telemetry and agentic risk

Overview

This skill is a local document checker that highlights sensitive words in user-selected files, with no evidence of hidden networking, credential access, persistence, or destructive intent.

Before installing, understand that this skill processes local documents and writes highlighted output files, usually to the Desktop unless another output folder is provided. Use an explicit output directory and avoid pointing it at the same location and filename as an original document if you want to preserve the unmarked file.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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)

Lp3

Medium
Category
MCP Least Privilege
Confidence
83% confidence
Finding
The skill instructs the agent to create annotated output files but does not declare any permissions for file writing. This creates a transparency and policy-enforcement gap: a host system or reviewer may underestimate the skill's ability to modify or generate files, increasing the chance of unauthorized writes or unsafe execution in broader workflows.

Vague Triggers

Medium
Confidence
78% confidence
Finding
The trigger keywords include broad terms like '检查' and '审核' without narrow scope constraints, which can cause the skill to activate for generic document-review tasks unrelated to sensitive-word scanning. Over-broad activation increases the chance of unnecessary file processing and unintended document modification/output in contexts where the user did not specifically request this behavior.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The skill description emphasizes checking files but does not clearly warn users that it creates annotated output copies and modifies document content presentation by highlighting matches. This can lead to surprise file changes, accidental distribution of marked-up documents, or unintended handling of sensitive files under the assumption of read-only analysis.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.