Daily AI Briefing(每日AI简报)

Security checks across malware telemetry and agentic risk

Overview

This skill appears to be a benign daily AI-news briefing workflow that uses public web sources and a small local Desktop archive.

Before installing, know that the skill will browse public AI news sources and create, read, and write files under ~/Desktop/daily_ai_briefing. Keep that folder limited to intended briefing files, and check for an existing same-date file if overwriting a prior draft matters.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Missing User Warnings

Low
Confidence
94% confidence
Finding
The skill instructs the agent to read from and write to a fixed local Desktop path (`~/Desktop/daily_ai_briefing/`) and to create the folder if it does not exist, without any explicit user consent or warning. This can cause unintended local file modification and disclosure of nearby user data in prior briefing files, even though the functionality appears intended for normal workflow automation rather than abuse.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal