ClawHub
Back to skill

Security audit

pixmind-video

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed PixMind video-generation helper that uses a PixMind API key to call PixMind-documented API endpoints.

Before installing, understand that your PixMind API key, prompts, and any reference image URLs will be sent to PixMind's API service and may incur API usage costs. Use a scoped/revocable key where possible.

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 (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
89% confidence
Finding
The skill clearly requires environment access to read `PIXMIND_API_KEY` and network access to call external endpoints, but those capabilities are not declared as permissions in a standardized way. This creates a transparency and governance gap: users or platforms may not realize the skill can exfiltrate prompts, image URLs, and use a secret-bearing API key to contact third-party infrastructure.

Description-Behavior Mismatch

Medium
Confidence
98% confidence
Finding
The skill claims to use Pixmind, but hardcodes requests to a different domain (`aihub-admin.aimix.pro`) while still asking the user to supply a `PIXMIND_API_KEY`. This creates a credential redirection risk: the user's API key and all prompts are sent to an unexpected backend, which could harvest secrets, proxy requests without authorization, or misuse billing on the user's account.

Intent-Code Divergence

Medium
Confidence
96% confidence
Finding
The help/error text explicitly instructs the user to provide a Pixmind API key, but the implementation sends the authenticated request to a different backend domain. This mismatch increases the likelihood of successful credential theft or undisclosed third-party data sharing because users are led to trust they are authenticating directly with Pixmind when they are not.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.