ClawHub
Back to skill

Security audit

Harmonic.ai

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed Harmonic.ai connector wrapper with limited tool scope and no evidence of hidden or destructive behavior.

Install this only if you are comfortable using OOMOL as the intermediary for Harmonic.ai actions. User queries and identifiers such as company websites, LinkedIn URLs, or email addresses may be sent to the connector/Harmonic.ai when you ask for enrichment or lookup.

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 (1)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The activation text says to use this skill for ANY Harmonic.ai request and whenever a task involves Harmonic.ai, which is broader than necessary and can cause the agent to invoke the skill on vague references rather than clear user intent. In a tool-enabled environment, over-broad routing increases the chance of unnecessary external data access or unintended disclosure of user queries to the Harmonic.ai connector.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal