ClawHub
Back to skill

Security audit

World News Aggregator Skill

Security checks across malware telemetry and agentic risk

Overview

This is a normal news aggregation skill with ordinary external feed access, but users should be aware it may install Python dependencies when its helper script runs.

Install this if you are comfortable with a skill that contacts third-party public news and feed sites. Run the helper script in a virtual environment or sandbox if you do not want automatic pip dependency installation to affect your main Python setup, and use explicit commands for sensitive research topics.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

subprocess module call

Medium
Category
Dangerous Code Execution
Content
except ImportError:
    print("Installing dependencies...")
    import subprocess
    subprocess.check_call([sys.executable, "-m", "pip", "install", "-r", "requirements.txt"])
    import requests
    import feedparser
Confidence
95% confidence
Finding
The script automatically installs dependencies at runtime by invoking pip if imports fail. In an agent or automation context, this creates a supply-chain and environment-modification risk: execution may pull and install whatever versions are resolved from requirements.txt or package indexes without prior review, changing the host state and potentially introducing malicious or compromised packages.

Vague Triggers

Medium
Confidence
86% confidence
Finding
The README includes very broad natural-language trigger examples such as requests for latest AI papers or new policy updates, which can overlap with ordinary conversation. In an agent skill system, overly generic trigger phrasing can cause unintended invocation of the skill, leading to unexpected network access, data retrieval, or misleading execution in contexts where the user did not explicitly intend to run the skill.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The README advertises multi-source aggregation, intelligent search, and AI summarization but does not clearly disclose that the skill may perform outbound network requests to numerous external sources and potentially transmit retrieved content for processing. In an agent environment, missing transparency around network behavior undermines informed consent and can expose user environments to unexpected data transfer, third-party contact, or policy violations.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.