ClawHub
Back to skill

Security audit

Seedream 5.0 Lite 文生图

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent AI image-generation helper that uses an AI Hive API key, sends text prompts to a fixed service, and downloads generated images locally.

Before installing, be aware that this skill sends your text prompts and API key-authenticated requests to AI Hive, can save an API key locally, and downloads generated files to your machine. Use a dedicated API key if possible and review the output directory and config file location if you share the device.

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
80% confidence
Finding
The skill advertises and demonstrates shell execution, network access, local file reads/writes, and environment-variable use, but no explicit permissions are declared. This creates a transparency and governance gap: users and hosting platforms may not realize the skill can store data locally, access credentials, and download files, increasing the chance of over-trusting or mis-scoping the skill.

Tp4

High
Category
MCP Tool Poisoning
Confidence
91% confidence
Finding
The description frames the skill narrowly as text-to-image generation without reading reference images, but the documented behavior also includes interactive API key setup, local credential storage, task querying, and downloading outputs to disk. This mismatch can mislead users about the trust boundary and data handling, causing them to invoke a skill with broader local and network effects than expected.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.