ClawHub

Fairytale Dream Transform Video Gen Seedance2.0

Security checks across malware telemetry and agentic risk

Overview

This appears to be a disclosed creative video-generation skill that sends user-provided prompts and image URLs to WeryAI, with no concrete evidence of hidden, destructive, or credential-stealing behavior.

Install only if you are comfortable sending the prompts and image URLs you provide to WeryAI for processing. Avoid submitting secrets, private media URLs, unreleased campaign assets, or regulated data unless that third-party use is approved. If you need strict model control, verify the skill or script enforces the intended Seedance 2.0 model rather than relying only on documentation.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Tp4

High
Category
MCP Tool Poisoning
Confidence
88% confidence
Finding
The skill claims to be locked to a single benign-seeming style/model, but the referenced CLI behavior reportedly supports arbitrary model IDs and broader WeryAI operations. That mismatch can mislead users and calling agents into trusting a narrowly scoped style tool when it is actually a general-purpose remote API client, increasing the risk of unintended network actions, policy bypass, or use of unreviewed models/endpoints.

Vague Triggers

Medium
Confidence
90% confidence
Finding
The implicit trigger test is broad enough that the skill may auto-invoke on loosely related requests that mention aesthetic terms like pastel, fairytale, or vertical. This can cause unintended skill activation, reducing routing precision and potentially overriding user intent or causing unexpected model behavior in contexts where the skill was not explicitly requested.

Missing User Warnings

Medium
Confidence
78% confidence
Finding
The documented commands send prompts and public image URLs to external WeryAI endpoints, but the file does not warn users that potentially sensitive content will leave the local environment and be processed by a third party. In a creative media skill, users may include private images, unpublished campaign content, or confidential text prompts, so lack of disclosure increases the chance of inadvertent data exposure.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal