Back to skill

Security audit

Taobao Ecommerce System

Security checks across malware telemetry and agentic risk

Overview

This appears to be a Taobao ecommerce helper rather than malware, but it advertises live order, listing, advertising, and customer-service automation without reliably scoped controls or matching implementation safeguards.

Treat this as a demo or template unless you can audit and constrain it. Do not connect real Taobao, 1688, advertising, order, or customer-service credentials until the publisher documents exact API scopes, adds explicit previews and confirmations for every external or state-changing action, and makes the output clearly distinguish simulated actions from real ones.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (14)

Vague Triggers

Medium
Confidence
92% confidence
Finding
The trigger phrase set includes broad natural-language expressions for a command that performs external data retrieval and analysis. Generic phrases can cause unintended invocation by the agent during ordinary conversation, especially when the user is discussing sourcing rather than explicitly requesting execution.

Vague Triggers

Medium
Confidence
98% confidence
Finding
A generic trigger for order processing is dangerous because this action can submit orders to external supplier platforms and may use customer/order data. In this skill's context, accidental invocation is more severe because it can initiate real-world transactions, affect third-party systems, and expose personal or operational data.

Vague Triggers

Medium
Confidence
95% confidence
Finding
The phrase for starting customer-service automation is too generic and can be matched during normal support discussions. If triggered unintentionally, the system could begin automated reply behavior without the operator realizing it, causing inappropriate or misleading communications.

Vague Triggers

Medium
Confidence
96% confidence
Finding
An overly broad trigger overlapping with common customer-service language increases the chance that routine conversation activates automation. In this e-commerce context, unintended auto-replies could affect customer trust, create compliance issues, and send messages the operator did not approve.

Missing User Warnings

High
Confidence
97% confidence
Finding
The order-processing section describes automatic pushing of Taobao orders to 1688/网商园 but does not warn that it may submit data to external systems or initiate real transactions. Because the feature handles customer addresses and supplier ordering workflows, missing warnings materially increases the risk of unauthorized actions and privacy-impacting mistakes.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The customer-service automation section omits a warning that enabling the feature may send automated replies to customers. In this context, undisclosed automation can lead to accidental outbound messaging, reputational harm, and potentially non-compliant communications if operators do not understand that replies may be sent automatically.

Vague Triggers

Medium
Confidence
94% confidence
Finding
The documented dialogue trigger is a very generic phrase ('智能选品') with no namespace, confirmation step, or exclusion rules. In an agent environment, broad natural-language triggers can cause unintended skill invocation during ordinary conversation, which is especially risky here because the skill can initiate business workflows such as product selection and downstream automation.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The file advertises automatic order pushing to 1688 and automatic customer-service replies without any warning, approval gate, audit note, or discussion of data handling. Because these functions can act on customer information, supplier contacts, and live business operations, lack of explicit risk disclosure and control guidance increases the chance of unauthorized actions, privacy mistakes, and financially harmful automation.

Unpinned Dependencies

Low
Category
Supply Chain
Content
# 版本:2.0.0

# HTTP 请求
requests>=2.31.0

# 数据处理
pandas>=2.0.0
Confidence
88% confidence
Finding
requests>=2.31.0

Unpinned Dependencies

Low
Category
Supply Chain
Content
requests>=2.31.0

# 数据处理
pandas>=2.0.0
openpyxl>=3.1.0

# 图像处理
Confidence
88% confidence
Finding
pandas>=2.0.0

Unpinned Dependencies

Low
Category
Supply Chain
Content
# 数据处理
pandas>=2.0.0
openpyxl>=3.1.0

# 图像处理
Pillow>=10.0.0
Confidence
88% confidence
Finding
openpyxl>=3.1.0

Unpinned Dependencies

Low
Category
Supply Chain
Content
openpyxl>=3.1.0

# 图像处理
Pillow>=10.0.0

# 环境变量加载
python-dotenv>=1.0.0
Confidence
91% confidence
Finding
Pillow>=10.0.0

Unpinned Dependencies

Low
Category
Supply Chain
Content
Pillow>=10.0.0

# 环境变量加载
python-dotenv>=1.0.0

# 日志
colorlog>=6.7.0
Confidence
84% confidence
Finding
python-dotenv>=1.0.0

Unpinned Dependencies

Low
Category
Supply Chain
Content
python-dotenv>=1.0.0

# 日志
colorlog>=6.7.0
Confidence
82% confidence
Finding
colorlog>=6.7.0

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.