Security audit

Etsy 电商图片生成与编辑

Security checks for vulnerabilities and agentic risk

Overview

This skill is an Etsy image-generation wrapper that uses a disclosed external AI service and user-provided files, with some unused generic client code that should be noted but is not exposed by the normal skill entry point.

Install only if you are comfortable sending selected listing/reference images and prompts to AI Hive and storing an AI Hive API key locally. Avoid using the upload command on unrelated private files, and review the provider account/billing implications of generated tasks.

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: 91% confidence
Finding: The skill invokes shell commands, accesses local files and environment variables, and installs/uses network-capable Python tooling, yet it declares no permissions. This creates a transparency and containment problem: a caller may expect a simple image-editing skill, while the implementation can read local inputs, write files, reach external services, and execute commands without explicit disclosure or gating.

Tp4

High

Category: MCP Tool Poisoning
Confidence: 96% confidence
Finding: The documented purpose is narrowly framed as Etsy listing image generation/editing, but the skill reportedly also supports broad AI Hive capabilities such as chat, video generation, account and wallet queries, model listing, key setup, and generic uploads. That mismatch is dangerous because it hides materially different data access and external interactions, increasing the risk of credential exposure, unauthorized API use, privacy leakage, and user surprise beyond the stated business purpose.

Description-Behavior Mismatch

Medium

Confidence: 95% confidence
Finding: The file implements a broad, generic AI client with chat, model enumeration, media upload, and video generation rather than a narrowly scoped Etsy image-editing skill. This scope mismatch increases attack surface and permits users or downstream agents to invoke unrelated capabilities, undermining least privilege and making misuse easier.

Context-Inappropriate Capability

Medium

Confidence: 93% confidence
Finding: Exposing a user-info endpoint in a skill intended for Etsy image generation is unnecessary privilege expansion. It allows retrieval of account and wallet details unrelated to the task, creating avoidable privacy and reconnaissance risk if the skill is invoked by an untrusted workflow or agent.

Context-Inappropriate Capability

Medium

Confidence: 94% confidence
Finding: General video-generation support is outside the declared image-editing purpose of the skill and materially broadens functionality. Extra media types, upload paths, and generation endpoints increase complexity and the chance of unintended use, abuse, or policy bypass through a mislabeled skill.

Intent-Code Divergence

Medium

Confidence: 88% confidence
Finding: The module advertises itself as a generic AI capability tool, which contradicts the declared specialized Etsy-image intent. This misleading packaging can cause operators or orchestration systems to trust and deploy a much more capable tool than expected, contributing to overprivileged execution.

Static analysis

No suspicious patterns detected.