ClawHub
Back to skill

Security audit

xiaohongshu-extract

Security checks across malware telemetry and agentic risk

Overview

This is a straightforward Xiaohongshu metadata extractor with normal web-request and optional local-file-output risks.

Install if you are comfortable with the agent making outbound requests for links you provide. Use public Xiaohongshu URLs, avoid private or internal URLs, and choose --output paths carefully because the script can write extracted metadata to that local path.

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
91% confidence
Finding
The skill documentation instructs running a bundled Python script that performs network access to user-supplied URLs and can write results to arbitrary output paths, yet the skill declares no permissions. This mismatch is a real security issue because it hides the true capability surface from the agent/user, reducing informed consent and increasing the chance of unsafe use of network and filesystem operations.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.