Security audit

微信小店电商视频生成与编辑

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent AI Hive video-generation helper, with some credential-storage and bundled unused generic client code that users should understand before installing.

Before installing, understand that this skill sends prompts and selected media to AI Hive, stores an API key locally if you run init, and saves generated outputs under ~/Downloads/AiHive unless changed. Use only media you are allowed to upload and avoid passing sensitive local files as references.

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
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

Findings (7)

Lp3

Medium

Category: MCP Least Privilege
Confidence: 85% confidence
Finding: The skill invokes a Python script that performs package installation, shell execution, file access, and likely network operations, yet the skill file declares no explicit permissions or capability boundaries. This creates a transparency and policy-enforcement gap: users and platforms cannot accurately assess or constrain what the skill may access when run.

Tp4

High

Category: MCP Tool Poisoning
Confidence: 92% confidence
Finding: The documented purpose is narrowly scoped to e-commerce video generation/editing, but the detected behavior includes unrelated capabilities such as wallet/user info queries, model enumeration, general chat, media upload, browser interaction, and API key initialization. This mismatch is dangerous because it can conceal overbroad access and enable data exposure, credential handling, or unintended actions outside the user's expected trust boundary.

Description-Behavior Mismatch

Medium

Confidence: 96% confidence
Finding: The file bundles a broad generic AI Hive client and CLI, including chat, image generation, model enumeration, media upload, and account-related endpoints, while the declared skill is specifically for WeChat-shop video generation/editing. This capability mismatch expands the attack surface and enables use of the skill for unintended actions that a caller or reviewer would not reasonably expect from the manifest.

Context-Inappropriate Capability

Medium

Confidence: 92% confidence
Finding: Exposing user/account information and wallet balance is not necessary for ecommerce video generation and creates unnecessary access to sensitive account metadata. In a skill context, this is dangerous because callers may grant or use the skill expecting only media generation, not account inspection.

Context-Inappropriate Capability

Medium

Confidence: 94% confidence
Finding: The standalone text chat endpoint is unrelated to the stated purpose of WeChat-shop video generation/editing and permits arbitrary prompt-based interaction with the backend. This broadens the skill beyond its declared scope and can enable unauthorized or unexpected use patterns under the guise of a narrow video tool.

Context-Inappropriate Capability

Medium

Confidence: 94% confidence
Finding: Standalone image generation is outside the declared skill purpose and provides an additional content-generation channel not implied by a video-editing skill. This mismatch matters because users and systems may rely on manifest scoping for trust and policy decisions.

Missing User Warnings

Medium

Confidence: 84% confidence
Finding: The init flow persists the API key to disk automatically after paste without an explicit warning that credentials will be stored locally. Although the file is permissioned to 0600, silent persistence can still surprise users and increase the risk of credential exposure on shared, backed-up, or compromised systems.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.