ClawHub
Back to skill

Security audit

Shopify REST Admin (Legacy)

Security checks across malware telemetry and agentic risk

Overview

This Shopify skill is mostly read-oriented, but it gives an agent a broad generic connector path that is not clearly limited to the documented read-only actions.

Review this skill before installing if your Shopify connection has write or delete permissions. Use it only for the listed read actions unless you have separately confirmed the exact connector action, payload, and effect, and be aware that setup may install and authenticate the OOMOL CLI.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Description-Behavior Mismatch

High
Confidence
97% confidence
Finding
The skill metadata says it should be used for 'searching and reading data', but the body explicitly discusses write and destructive actions. That mismatch can cause downstream agents or users to trust the skill as read-only and invoke it in contexts where mutation should be disallowed, enabling unintended state changes in Shopify if additional actions exist or are later added.

Intent-Code Divergence

Medium
Confidence
93% confidence
Finding
The instructions tell the agent to inspect any action schema and run any action name via a generic connector interface, while the safety section implies safety based only on the listed actions. If the connector exposes additional unlisted state-changing actions, an agent could execute them because the invocation pattern is unrestricted and the safety assumptions are incomplete.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.