Intern Report

Security checks across malware telemetry and agentic risk

Overview

This is a local internship-report generator; one extra helper script keeps a local command history, but I found no network, credential, destructive, or hidden automatic behavior.

Use the documented scripts/intern.sh commands for report generation. Avoid passing secrets, private file paths, or sensitive personal data to scripts/script.sh unless you are comfortable with those arguments being stored locally in history.log.

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
90% confidence
Finding
The script persistently logs command usage and user-supplied arguments to a file under the user's data directory without clearly disclosing that behavior. This can expose sensitive filenames, query terms, export destinations, or other operational details to anyone with access to the account or collected user data, especially because multiple commands pass raw arguments into the log.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal