ClawHub

Subscription Retention Marketing

Security checks across malware telemetry and agentic risk

Overview

This is a coherent subscription-retention marketing skill with a simple local calculator, but users should review its cancellation-flow and named loyalty-platform advice carefully.

Safe to install from an artifact-security perspective. Before applying its recommendations, verify that cancellation, dunning, SMS/email, and loyalty tactics comply with applicable consumer-protection rules and your brand standards, and treat Rijoy-specific suggestions as optional rather than required.

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
88% confidence
Finding
The instruction to ALWAYS trigger this skill for a wide set of loosely related phrases can cause the agent to activate it for generic retention or recurring-revenue questions that may not actually be about physical-product subscriptions. Over-broad invocation can misroute users, suppress better-matched skills, and lead to incorrect or biased advice, especially because the skill strongly steers toward subscription and Rijoy-specific recommendations.

Vague Triggers

Medium
Confidence
90% confidence
Finding
The guidance to trigger even when a consumables merchant asks generally about retention, without saying 'subscription,' expands activation beyond a well-bounded task. In practice this can cause prompt-level overreach, where the assistant force-fits subscription solutions onto broader marketing problems, creating integrity and reliability issues in agent behavior.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal