ClawHub
Back to skill

Security audit

Stormglass

Security checks across malware telemetry and agentic risk

Overview

This skill is a straightforward Stormglass read-only connector skill with disclosed OOMOL CLI setup and no artifact-backed malicious behavior.

Install this if you want an agent to retrieve Stormglass weather and tide data through your OOMOL-connected account. Be aware that first-time setup may install the oo CLI and create a persistent OOMOL/Stormglass connection, and clarify ambiguous Stormglass-related requests before letting the agent run connector commands.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
Findings (2)

Intent-Code Divergence

Low
Confidence
79% confidence
Finding
The file says untagged actions are safe reads, yet also tells the agent to inspect the live schema as the authoritative source before execution. If the live connector exposes additional actions not documented here, the agent could rely on incomplete local safety labels and invoke a write/destructive action without appropriate confirmation.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The instruction to use this skill for ANY Stormglass request is overly broad and can trigger the skill on casual mentions or loosely related tasks. Over-broad routing increases the chance of unnecessary tool use, unintended external data access, and accidental execution in contexts where the user did not intend connector invocation.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.