ClawHub
Back to skill

Security audit

News Sentiment免费版

Security checks across malware telemetry and agentic risk

Overview

This skill is a basic financial news sentiment guide with disclosed public-news data sources and no hidden executable files, though its callback and execution instructions are loose.

Install only if you are comfortable with the agent fetching market/news data from external sources. Do not provide a callback URL unless you intentionally want results sent there, and confirm the stock ticker, time window, market, and any command execution before running the workflow.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

Medium
Confidence
89% confidence
Finding
The skill description says it is driven by natural-language instructions and is suitable for broad development scenarios, which makes invocation boundaries unclear. In an agent environment, vague activation criteria can cause the skill to run in contexts the user did not explicitly intend, increasing the chance of unintended network access or code execution via the declared exec capability.

Vague Triggers

Medium
Confidence
93% confidence
Finding
The usage flow instructs users to invoke the skill by generally describing a task in natural language, rather than through a constrained command or schema. For a skill with read and exec tools, this broad activation model increases the risk of accidental invocation, prompt-trigger confusion, and unintended execution paths initiated from ambiguous conversation context.

Missing User Warnings

Medium
Confidence
84% confidence
Finding
The skill accepts a callback_url and performs external data collection from Yahoo Finance and Google News, but it does not clearly warn users that inputs, results, or metadata may be transmitted over the network. In agent settings, this can lead to privacy surprises, data exfiltration to attacker-controlled callback endpoints, or disclosure of sensitive watchlists and workflow details.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.