ClawHub
Back to skill

Security audit

小红书 电商图片生成与编辑

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed AI Hive image-generation helper for Xiaohongshu ecommerce visuals, with expected API-key setup, user-selected file upload, network calls, and output downloads.

Install only if you are comfortable using AI Hive: the skill stores or reads an API key, sends your prompts and selected reference files to the AI Hive service, and saves generated outputs locally. Avoid passing private or sensitive images unless you intend to upload them to that provider.

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 executable workflows that install packages, run Python, access environment variables, read/write files, and make network requests, yet it declares no permissions. This creates a transparency and governance gap: users or orchestrators may invoke a capability-rich skill without informed consent or policy controls, increasing the risk of unintended data access or external exfiltration through the referenced script.

Tp4

High
Category
MCP Tool Poisoning
Confidence
94% confidence
Finding
The documented purpose is limited to Xiaohongshu ecommerce image generation/editing, but the detected behavior includes unrelated and more powerful capabilities such as chat/text APIs, video generation, generic media upload, model enumeration, user-info/wallet queries, and interactive browser/API-key setup. This mismatch is dangerous because it can mislead users and reviewers about the true attack surface, enabling unnecessary collection of account metadata, credential handling, and broader external service interaction than the description justifies.

Description-Behavior Mismatch

High
Confidence
98% confidence
Finding
The file is packaged as a Xiaohongshu ecommerce image skill, but it embeds a broader AI Hive client exposing unrelated capabilities such as text chat, video generation, user-info lookup, model enumeration, and arbitrary media upload. This expands the effective privilege and attack surface beyond the declared purpose, enabling misuse of the skill as a generic API wrapper and increasing the chance of unauthorized data access or policy bypass.

Intent-Code Divergence

Medium
Confidence
90% confidence
Finding
The module docstring explicitly describes a general-purpose AI capability tool, which contradicts the declared narrow skill purpose. This mismatch is dangerous because reviewers and users may trust the metadata while the implementation still contains broader functionality, undermining transparency and consent about what the skill can actually do.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.