ClawHub

SZZG007 Facebook Engage

Security checks across malware telemetry and agentic risk

Overview

This skill openly automates Facebook follows and public comments, but its scope is loose and includes guidance for avoiding bans and mimicking human behavior.

Install only if you are comfortable letting an agent use your logged-in Facebook browser session to follow pages and publish comments. Require manual review of the target page and exact comment before each action, and avoid using it for unsolicited, repeated, or platform-policy-violating engagement.

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
96% confidence
Finding
The trigger conditions are broad enough to match ordinary requests about social media interaction, which increases the chance of unintended activation of a skill that performs real external actions. In this context, the skill automates following and commenting on Facebook, so accidental triggering could cause unauthorized account activity, spam-like behavior, or platform policy violations.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The description omits clear warning that the skill will perform account actions on a third-party platform by auto-following and posting comments. That lack of disclosure is risky because users may not realize the skill can create public-facing content and alter account state, which is especially sensitive in a social-media automation context that may violate platform rules or user expectations.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal