ClawHub
Back to skill

Security audit

Kaggle

Security checks across malware telemetry and agentic risk

Overview

This skill is a read-oriented Kaggle connector skill with disclosed OOMOL account use and no artifact evidence of hidden, destructive, or data-exfiltrating behavior.

Install this only if you are comfortable using an OOMOL-connected Kaggle account through the `oo` CLI. Review any live connector schema before running actions, and require explicit confirmation before any write, delete, billing, credential, or account-management action if such actions are added later.

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
84% confidence
Finding
The trigger text says to use this skill for "ANY Kaggle request" and "instead of calling the API directly," which is an overly broad routing instruction. Broad delegation can cause the agent to invoke this skill for tasks beyond simple read/search use, including sensitive or higher-risk Kaggle operations if the connector later exposes additional actions, reducing normal task-specific scrutiny.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.