Publish Hsciq Mcp

• Outcome: Review
• Latest audit: Aug 31, 2026 · 1:04 PM UTC
• Version: 3.0.1

Overview

This skill appears purpose-built for customs-code lookup, but it can automatically upload product details and images to an external review service without a clear confirmation step.

Review before installing if your product photos, specifications, supplier details, brand/model data, QQ or WeChat contact information, or customs questions are confidential. Use this only when you intend to send those materials to HSCIQ for processing and possible human review, and require the agent to ask for confirmation before any create_guilei_form or discussion-posting action.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Rogue Agent: Self-Modification, Session Persistence
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (6)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 91% confidence
• Finding: The skill declares environment-variable use, local file access for image upload, and outbound network access, but it does not explicitly declare permissions. This weakens transparency and consent boundaries for users and platforms, making it easier for the skill to access secrets and transmit local data without clear policy enforcement.

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: The README states that creating a classification consultation will automatically submit product information and at least one product image to an external HSCIQ service for AI processing and expert review, but it does not prominently warn users that potentially sensitive commercial data will leave their environment. In a business/customs context, product specs, images, and declaration details may contain confidential trade information, so silent or under-disclosed transmission creates a real privacy and data-handling risk.

Missing User Warnings

Medium

• Confidence: 88% confidence
• Finding: The skill supports submitting product information and images to an external HSCIQ service for expert review, but the surrounding guidance does not clearly warn that user-supplied data will leave the local environment. This creates a privacy and data-handling risk, especially if users provide sensitive product, supplier, or prototype images.

Missing User Warnings

High

• Confidence: 97% confidence
• Finding: The example explicitly says the AI will automatically submit product information and images for manual review when uncertain, without requiring a confirmation step. That is dangerous because it normalizes autonomous exfiltration of potentially sensitive commercial data to a third party, including images that may contain confidential product designs or labels.

Missing User Warnings

Medium

• Confidence: 89% confidence
• Finding: The CLI accepts product details, contact information, and local image files, then immediately transmits them to a remote API when the command is run, without an inline confirmation prompt or a clear just-in-time disclosure. This creates a real privacy and data-handling risk because users may unintentionally upload sensitive business information or personal contact data, especially in an agent or automation context where command execution may be less visible.

Session Persistence

Medium

• Category: Rogue Agent

Content

---
name: hsciq-mcp
description: HS Code Lookup for Chinese Products. Query customs codes, tariff rates, declaration elements, and regulatory requirements via HSCIQ MCP API. Create classification consultation requests with image upload for expert review.
license: MIT
tags:
  - HS Code Lookup for Chinese Products

• Confidence: 70% confidence
• Finding: The skill creates and later retrieves persistent consultation forms and associated field discussions on a remote service. That persistence is expected for the business function, but it still introduces retention and exposure risk because uploaded images, product details, and conversation history may remain stored server-side beyond the immediate session.

VirusTotal

VirusTotal findings are pending for this skill version.

Static analysis

Detected: suspicious.env_credential_access

Environment variable access combined with network send.

Critical

• Code: suspicious.env_credential_access
• Location: hsciq-client.js:10