ClawHub
Back to skill

Security audit

知乎内容搜索分析

Security checks across malware telemetry and agentic risk

Overview

This skill is a straightforward Zhihu search and topic-analysis helper that makes disclosed, purpose-aligned requests to Zhihu and shows no hidden access or persistence.

Install only if you are comfortable running a local Python script that sends your Zhihu search terms to Zhihu. Avoid sensitive queries, expect network access to Zhihu endpoints, and note that VirusTotal was still pending while static and artifact review found no malicious behavior.

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
89% confidence
Finding
The skill explicitly instructs users to run a Python script that performs Zhihu searches, hot-list retrieval, and question analysis via a networked API, yet the skill declares no permissions. This creates a transparency and governance gap: users or platforms may authorize or trust the skill without understanding it makes outbound network requests, which can affect privacy, compliance, and reviewability.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal