ClawHub
Back to skill

Security audit

Seedream 5.0 Lite 直播带货图片

Security checks across malware telemetry and agentic risk

Overview

This skill is a purpose-aligned AI Hive image-generation helper, with expected API key setup, image upload, task polling, and local result saving behavior.

Before installing, understand that this skill uses an AI Hive API key, may upload your chosen reference images to AI Hive/object storage, and can save the key under ~/.ai-hive/config.json. Use only approved images and prompts, and remove or rotate the stored key if the machine is shared.

SkillSpector

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
93% confidence
Finding
The skill exposes significant capabilities—shell execution, network access, environment access, and file read/write—without any declared permissions or clear capability boundaries. That makes the trust model opaque and increases the chance that users or orchestrators invoke a skill with broader access than its stated purpose suggests, enabling credential handling, local file modification, or unexpected outbound requests.

Tp4

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding
The documented behavior says the skill is for narrowly scoped livestream image generation, but the described commands support broader actions such as interactive credential initialization, persistent config writes under the user's home directory, arbitrary task lookup, and reusable media upload. This mismatch is dangerous because it can mislead users and policy systems into granting trust to a skill that performs more sensitive operations than advertised, especially around credential handling and persistent local changes.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.