ClawHub
Back to skill

Security audit

Seedream 5.0 Lite 产品图生成

Security checks across malware telemetry and agentic risk

Overview

This skill is mainly a disclosed AI Hive product-image generator, with expected API key, upload, and download behavior, though its bundled helper code is broader than the user-facing skill description.

Install only if you are comfortable giving the skill an AI Hive API key, uploading chosen product/reference media to AI Hive or its object storage, and saving generated results locally. Avoid pointing the upload or generate commands at private files that are not intended for that service.

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 (4)

Lp3

Medium
Category
MCP Least Privilege
Confidence
87% confidence
Finding
The skill advertises shell execution, file access, environment access, and network use without declaring permissions, which undermines least-privilege expectations and makes the operational risk opaque to users and reviewers. In this context, the commands install packages and run a Python helper that can read/write local data and contact external services, so undeclared capabilities could enable unintended data exposure or unsafe execution paths.

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding
The documented purpose is narrow product-image generation, but the underlying behavior reportedly includes browser-based credential setup, local credential storage, account and wallet queries, model enumeration, text chat, video generation, uploads, and task querying. This mismatch is dangerous because users may invoke the skill expecting constrained image generation while the backing code exposes broader account, data-handling, and external-service functionality that increases attack surface and the chance of credential misuse or unexpected actions.

Description-Behavior Mismatch

Medium
Confidence
90% confidence
Finding
The skill is presented as a narrowly scoped Seedream 5.0 Lite product-image generator, but it also exposes generic task querying and arbitrary media upload operations. In an agent or marketplace context, this scope mismatch can bypass user expectations, policy gating, or tool allowlisting by granting broader API interaction than the metadata suggests.

Intent-Code Divergence

High
Confidence
95% confidence
Finding
The top-level module is a general-purpose AI Hive client supporting chat, image, video, model listing, upload, and polling, which materially exceeds the advertised Seedream 5 Lite product-image purpose. This kind of hidden capability expansion is dangerous because downstream systems may trust the skill metadata for permissioning while the actual code enables much broader remote API actions.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.