ClawHub
Back to skill

Security audit

小红书爆款笔记封面生成与编辑

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed AI image-cover generation skill that uses an external API, uploads user-selected reference images, downloads results, and stores an API key for future use.

Install this only if you are comfortable sending selected reference images and prompts to AI Hive and storing an AI Hive API key locally. Use dedicated API keys when possible, review provider billing and content rules, and avoid uploading private or sensitive images unless that is intended.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
85% confidence
Finding
The skill documentation invokes shell commands, installs dependencies, accesses local files via image paths, writes data locally during initialization, and performs network operations, yet it declares no permissions. This creates a transparency and consent problem: users may run a skill believing it is low-risk content generation while it can read/write locally and communicate externally.

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding
The stated purpose is narrow image-cover generation, but the skill reportedly also supports unrelated capabilities such as chat, video generation, wallet/user queries, model enumeration, broad media upload, and interactive API-key setup with opening an external browser. This mismatch is dangerous because it expands the attack surface far beyond user expectations, enabling unnecessary data access, credential handling, and external communications in a context where users may not scrutinize those behaviors.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.