Security audit

SHEIN 电商视频生成与编辑

Security checks for vulnerabilities and agentic risk

Overview

This skill is a coherent AI Hive video-generation helper for SHEIN-style ecommerce media, with expected API key, upload, polling, and download behavior.

Install only if you are comfortable giving this skill an AI Hive API key and uploading selected media to AI Hive/object storage for generation. Use only intended media files, review the configured base URL, and be aware that init saves the API key locally under ~/.ai-hive/config.json.

SkillSpector

Vulnerability Patterns

Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
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

Findings (6)

Lp3

Medium

Category: MCP Least Privilege
Confidence: 87% confidence
Finding: The skill documentation invokes a Python script that can install packages, access environment variables, read/write files, use the network, and execute shell commands, yet no permissions are declared. This creates a trust and containment gap: users and platforms cannot accurately assess or sandbox the skill’s actual capabilities, increasing the risk of unintended data access, exfiltration, or filesystem modification when the script runs.

Tp4

High

Category: MCP Tool Poisoning
Confidence: 91% confidence
Finding: The documented purpose is narrowly scoped to SHEIN ecommerce video generation/editing, but the associated behavior reportedly includes generic chat, image generation, account/wallet retrieval, model discovery, uploads, and task querying beyond that use case. This mismatch is dangerous because users may authorize or run the skill under false assumptions, while the broader functionality could expose account data, consume paid resources, or be repurposed for unrelated networked operations.

Description-Behavior Mismatch

Medium

Confidence: 96% confidence
Finding: The file is a broad AI Hive wrapper that exposes chat, image generation, model enumeration, media upload, user-info lookup, and task polling, while the declared skill is narrowly framed as SHEIN video generation/editing. This scope mismatch increases the attack surface and enables capability abuse outside the reviewed or expected business purpose, which is especially risky in agent ecosystems where users may trust the declared skill description.

Context-Inappropriate Capability

Low

Confidence: 88% confidence
Finding: The skill exposes a user/account information retrieval endpoint that is not necessary for creating or editing SHEIN videos. Even if limited to the current account, unnecessary account-data access broadens the data exposure surface and can reveal wallet, profile, or tenant information to an agent flow that users did not intend to authorize for this purpose.

Context-Inappropriate Capability

Medium

Confidence: 95% confidence
Finding: The generic text chat capability is unrelated to the stated SHEIN video-editing purpose and allows the skill to act as a general LLM proxy. In an agent setting, this can be used to exfiltrate prompts, process unrelated sensitive text, or bypass product-level restrictions by piggybacking on a skill that appears to be only for video work.

Context-Inappropriate Capability

Medium

Confidence: 94% confidence
Finding: Image generation is outside the declared scope of a SHEIN video generation/editing skill and materially expands what the tool can do. This increases misuse potential and may let downstream agents create content or process inputs that were never intended to be in scope for this integration.

Static analysis

No suspicious patterns detected.