ClawHub
Back to skill

Security audit

Higgsfield AI

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent Higgsfield AI connector that uses the OOMOL CLI and discloses its write actions and setup steps.

Before installing, be aware that the skill can submit and cancel Higgsfield AI generation jobs through your connected OOMOL account. Confirm prompts, request IDs, and cancellation targets carefully, and only run the CLI installer or account-connection steps if you actually need first-time setup.

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 trigger text says to use this skill for ANY Higgsfield AI request and whenever a task involves Higgsfield AI, which is overly broad and can cause the agent to invoke the skill for loosely related mentions rather than explicit user intent. In a skill that can perform write actions such as submitting or canceling generation requests, this increases the chance of unnecessary external calls or unintended state-changing operations.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.