ClawHub
Back to skill

Security audit

IT之家 日榜/周榜/月榜 热门文章

Security checks across malware telemetry and agentic risk

Overview

This skill openly fetches IT Home ranking articles and can send or schedule them to a user-chosen messaging channel.

Install this only if you are comfortable letting the agent fetch IT之家 pages and, when you choose, send the formatted results to your selected messaging channel or create a scheduled push. Review the target channel and cron settings before running the QwenPaw examples.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Lp3

Medium
Category
MCP Least Privilege
Confidence
88% confidence
Finding
The skill instructs the agent to create and run Python code that performs outbound network access to ithome.com, but the skill declares no corresponding permissions. This creates a policy gap: users or operators may believe the skill is constrained when it can actually fetch remote content and use that content in downstream messaging workflows, increasing the risk of unauthorized data egress or unexpected external interactions.

VirusTotal

52/52 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.