Supply Chain Optimization Amazon Lite

• Outcome: Pass
• Latest audit: May 29, 2026, 8:27 PM
• Version:
  0.1.0

Overview

This skill is a local supply-chain calculator with optional local reports, and the reviewed artifacts do not show hidden access, credential use, exfiltration, or destructive behavior.

Install only if you are comfortable entering supply-chain and margin figures into a local calculator. Choose text output if you do not want files saved, and treat generated Markdown or HTML reports as sensitive business documents; opening the HTML report may contact jsdelivr to load Chart.js.

SkillSpector

By NVIDIA

Vulnerability Patterns

• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (4)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 83% confidence
• Finding: The skill documentation claims a conversation-only tool with no declared permissions, yet it describes capabilities that can write generated reports to local storage. Undeclared file-write behavior creates a transparency and trust gap: users and the platform cannot accurately assess what data may be persisted, especially since the collected inputs include business-sensitive supply chain and financial information.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 91% confidence
• Finding: The stated purpose is conversational diagnosis, but the documented behavior expands into generating HTML/Markdown reports and writing them to disk. This mismatch is dangerous because it hides materially different data-handling behavior from users and reviewers, increasing the chance that sensitive business data is persisted or shared unexpectedly.

Description-Behavior Mismatch

Medium

• Confidence: 80% confidence
• Finding: The skill says no integration is required and presents itself as conversation-based, but the workflow instructs execution of an external script to process user-supplied data. Even if local, this is still code execution beyond pure conversation and changes the risk profile because calculation logic, input handling, and side effects are no longer transparent in the top-level documentation.

Missing User Warnings

Medium

• Confidence: 88% confidence
• Finding: The skill offers web-report and document-generation options without warning users that their supply chain, pricing, and margin data may be transformed into persistent artifacts. In this business context, those reports can contain commercially sensitive information, so silent persistence or shareable output creation increases confidentiality and data-handling risk.

VirusTotal

66/66 vendors flagged this skill as clean.