ClawHub

Notification Hub

Security checks across malware telemetry and agentic risk

Overview

This is an instruction-only notification hub that clearly describes collecting local event files, sending prioritized alerts, and keeping notification history.

Before installing, confirm which skills can write to events/, what content is allowed in notifications, where Discord messages are sent, and how long memory/notifications/ history should be retained. Avoid putting sensitive message bodies in events unless you are comfortable with them appearing in summaries or Discord alerts.

SkillSpector

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

Vague Triggers

Medium
Confidence
94% confidence
Finding
The trigger phrases are broad natural-language terms like "notification" and "anything new," which are likely to match ordinary conversation and invoke the skill unintentionally. In a notification hub that aggregates alerts from multiple skills, accidental activation can expose sensitive summaries such as DMs, security events, or operational status to the wrong context or user flow.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill explicitly centralizes and reads event files from multiple sources, including Instagram DMs, security alerts, and system health data, but does not warn users that sensitive content may be aggregated and surfaced together. Without clear disclosure and handling expectations, users may unintentionally expose private messages, security incidents, or operational metadata through summaries, heartbeat responses, or notification history views.

VirusTotal

61/61 vendors flagged this skill as clean.

View on VirusTotal