ClawHub
Back to skill

Security audit

oled-intelligence-portal

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed report generator that searches authorized Patsnap sources and writes local HTML pages, with some containment and input-sanitization caveats users should understand.

Install only if you intend to use Patsnap MCP-backed searches and generate local HTML reports. Review the generated portal data before relying on it, and avoid passing untrusted raw HTML or unusual path-like keywords/slugs because the renderer does not show escaping or path sanitization.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
86% confidence
Finding
The skill describes use of environment access and filesystem read/write behavior without declaring corresponding permissions. This creates a transparency and containment problem: reviewers and users cannot accurately assess what resources the skill may access or modify, and downstream platforms may enforce weaker controls than intended. In this context, the skill writes multiple HTML files and may rely on configuration data, so undeclared capabilities increase the risk of unexpected file creation or access.

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding
The skill's documented behavior claims autonomous company discovery, news retrieval, patent querying, and keyword expansion, but the implementation apparently depends on externally supplied JSON and does not perform those operations itself. This mismatch is dangerous because users may trust the generated portal as grounded in live external data when it may actually reflect unverified or injected input, enabling misinformation, spoofed intelligence outputs, or unsafe downstream decisions. The security risk is amplified because the skill produces polished HTML that can lend false credibility to fabricated content.

Vague Triggers

Medium
Confidence
82% confidence
Finding
Overly broad trigger phrases can cause the skill to activate on common requests unrelated to the user's true intent. In a skill that writes files and generates polished intelligence artifacts, misfires can lead to unintended tool use, unnecessary data retrieval, or creation of misleading outputs from ambiguous prompts. The context makes this more concerning because phrases like '帮我监控XXX技术' are broad and likely to overlap with normal research requests.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.