ClawHub

舆情监测智能体V2

Security checks across malware telemetry and agentic risk

Overview

The skill largely does what it claims, but it builds and shares an HTML news report from web content without clear safeguards around HTML/link sanitization, file output, or Feishu delivery control.

Install only if you are comfortable with the skill searching the web, creating a local HTML report, producing stock-related analysis, and sending that report to Feishu. Review generated reports before sharing, and prefer a version that escapes web-sourced text, validates links, makes the output path configurable, and confirms the Feishu recipient.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
86% confidence
Finding
The skill explicitly describes generating an H5 page via generate_html.py, which implies file creation/write behavior, yet no permissions are declared. Undeclared file-write capability weakens transparency and reviewability, making it easier for a skill to write unexpected files or overwrite local artifacts without clear operator awareness.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill sends generated content to Feishu and produces an H5 page, but it provides no user-facing notice, consent step, or data-handling warning before transmitting results externally. Even if the intended content is public news, generated summaries, links, analysis, and possibly metadata may be delivered to third-party systems without explicit user awareness, creating privacy, compliance, and unintended disclosure risks.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal