ClawHub

自动化营销推广师

Security checks across malware telemetry and agentic risk

Overview

This skill is transparent about automating marketing outreach, but it would monitor social platforms and automatically post or message people without clear limits or approval controls.

Install only if you will constrain it before use: require manual approval for posts and private messages, use least-privilege platform credentials, set rate and recipient limits, avoid personal-account bots where prohibited, and define what data is logged, how long it is kept, and how automation can be stopped.

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 (3)

Vague Triggers

Medium
Confidence
93% confidence
Finding
The monitoring keywords are very broad and cover common discussion topics across multiple public platforms, which can cause the agent to trigger on unrelated or low-intent conversations. In this skill, broad matching is directly tied to automated response generation and publication, increasing the risk of unsolicited outreach, spam, and reputational damage.

Vague Triggers

Medium
Confidence
96% confidence
Finding
The call rules state that detecting a topic category is enough to automatically publish a response, but they do not define decision boundaries, consent requirements, or exclusions. Because the skill is designed for unattended cross-platform posting and messaging, this creates a concrete risk of mass unsolicited posting, policy violations, and accidental engagement with inappropriate targets.

Missing User Warnings

High
Confidence
98% confidence
Finding
The skill explicitly describes automated monitoring, logging, auto-replying, and private messaging, yet provides no privacy notice, retention limits, consent model, or warning about platform and personal data impacts. In context, the tool processes public and potentially semi-private communications at scale, which raises compliance, privacy, and surveillance concerns beyond ordinary automation.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal