AI每日新闻 TOP 10

Security checks across malware telemetry and agentic risk

Overview

This skill simply opens a public AI news page and returns ten current headlines, with no credential, local-file, persistence, or mutation behavior found.

Install if you are comfortable letting your agent open ai-bot.cn in a browser to fetch public AI news headlines. No credentials or local data are requested, but the result depends on that website's availability and page content.

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
87% confidence
Finding
The trigger condition is broadly phrased to activate whenever a user asks about AI news or recent AI trends, without clear boundaries on when this skill should or should not take over. This can cause overbroad routing and unintended web access, making the agent invoke the skill in contexts where a simpler or safer response would have been sufficient.

VirusTotal

No VirusTotal findings

View on VirusTotal