ClawHub
Back to skill

Security audit

AI Short Video Script Generator Pro

Security checks across malware telemetry and agentic risk

Overview

The skill mostly matches its script-generator purpose, but its billing flow is unclear and overly broad enough that users should review it before installing.

Review the payment behavior before installing: the displayed price, runtime charge amount, and charge timing do not line up. Avoid entering confidential topics unless you accept processing by the external AI provider, and run the app only on a trusted local machine until the publisher removes hard-coded secrets, disables debug network exposure, and clarifies billing and data handling.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Description-Behavior Mismatch

Medium
Confidence
96% confidence
Finding
The manifest advertises $0.005 USDT per use, but the code sends billing requests for 0.001. Price mismatches in payment code are dangerous because they undermine informed consent and can lead to undercharging or overcharging if the values drift further across systems. In a paid skill, billing integrity is security-relevant because users and platforms rely on accurate, auditable pricing.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The skill explicitly advertises save, delete, and export functionality without warning users about data modification, permanence, or loss. Destructive operations on user-created scripts can lead to accidental deletion or overwrite, especially if there are no confirmation prompts, undo support, or retention guarantees. In this context, the danger is elevated because the stored content may represent paid outputs that users expect to preserve.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
User-provided topic content is transmitted to a third-party AI provider without any visible disclosure or consent mechanism in this file. This creates a privacy and data-governance risk because users may provide sensitive business ideas, campaign plans, or personal information assuming it stays within the skill.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The before_request hook attempts to charge users automatically before most requests, with no explicit confirmation step tied to each billable action. This is dangerous because a simple page interaction or API call can trigger billing unexpectedly, increasing the risk of unauthorized or surprise charges and making abuse easier if requests are replayed or induced.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.