Back to skill

Security audit

Etsy 电商视频生成与编辑

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed AI Hive video-generation helper for Etsy listing media, with expected network, credential, upload, and output-file behavior under user control.

Install only if you are comfortable giving this skill an AI Hive API key and uploading selected product photos, videos, or audio to AI Hive for processing. Review the media files before use, avoid including buyer personal data before order confirmation, and keep the API key file protected.

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 (5)

Lp3

Medium
Category
MCP Least Privilege
Confidence
87% confidence
Finding
The skill exposes shell, file, environment, network, and write capabilities through documented commands, yet no explicit permissions are declared. This creates a transparency and policy-enforcement gap: a user or host may assume the skill is low-risk while it can install packages, access local files, and communicate externally. In this context, the risk is elevated because the skill invokes a general Python helper that appears able to route tasks to external AI services and handle local media inputs.

Tp4

High
Category
MCP Tool Poisoning
Confidence
94% confidence
Finding
The documented purpose is Etsy video generation/editing, but the underlying behavior reportedly includes broad AI Hive functions such as generic chat, model enumeration, account/wallet retrieval, media upload, and initialization/configuration workflows. This overbroad capability set is dangerous because it can enable unexpected data exfiltration, account reconnaissance, or misuse of user credentials/tokens beyond the user’s reasonable expectations for a niche media-editing skill. The mismatch makes informed consent and least-privilege review difficult.

Description-Behavior Mismatch

Medium
Confidence
92% confidence
Finding
The file is a broad AI Hive client that supports text chat, image generation, model enumeration, uploads, and task polling, while the skill is presented as an Etsy video-generation/editing skill. This scope mismatch increases the attack surface and enables capabilities unrelated to the declared purpose, violating least-functionality expectations for agent skills.

Context-Inappropriate Capability

Medium
Confidence
88% confidence
Finding
The skill exposes account lookup and model discovery functions that are not necessary for Etsy listing video creation. In an agent environment, such surplus capabilities can reveal account metadata, available backend capabilities, and facilitate misuse beyond the user-expected workflow.

Context-Inappropriate Capability

Medium
Confidence
95% confidence
Finding
The generic text-chat command is unrelated to the stated Etsy video-generation/editing function and materially broadens what the skill can do. This creates an unnecessary prompt-execution channel that could be used for unrelated data processing or policy bypass within a skill that users would reasonably assume is video-only.

Static analysis

No suspicious patterns detected.