ClawHub
Back to skill

Security audit

Shopify Theme Apps Detector

Security checks across malware telemetry and agentic risk

Overview

This skill coherently scans public Shopify storefronts and creates a local report, with a privacy caveat that opened reports load third-party logos and favicons.

Install if you are comfortable scanning public Shopify stores and creating a local report. For sensitive audits, open reports with network access disabled or remove remote logo/favicon loading, because Google, Clearbit, and target/vendor domains may learn which report subjects were viewed.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Description-Behavior Mismatch

Low
Confidence
89% confidence
Finding
The template causes generated reports to contact third-party services such as Clearbit and Google to fetch logos/favicons. That leaks the analyzed store/app domains, reader IP/address metadata, and report-open timing to external parties, which is unnecessary for the core detection function and creates a privacy/data-disclosure issue.

Context-Inappropriate Capability

Low
Confidence
91% confidence
Finding
The report embeds tracking-capable third-party requests for cosmetic assets, so simply opening the report triggers outbound requests to external domains. This enables passive tracking of report viewers and leaks which stores/apps were analyzed, without being required for the report's main purpose.

Missing User Warnings

Low
Confidence
88% confidence
Finding
The skill explicitly instructs the agent to write an HTML report into the current working directory, which modifies the local filesystem without any user-facing warning or consent step. While this is expected behavior for a reporting skill and appears non-malicious, writing files to an implicit location can surprise users, overwrite existing files with similar names, or place output in sensitive/project directories.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The template performs undisclosed external network requests when the report is opened. In a security/audit context, that is more sensitive than a normal webpage because report viewers may assume the artifact is self-contained; instead it leaks access metadata and analyzed targets to Google, Clearbit, and the target domain itself.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.