Moin

Security checks across malware telemetry and agentic risk

Overview

This is an instruction-only tweet-writing skill whose web research step is disclosed and aligned with its purpose, but users should avoid giving it confidential campaign details.

Safe to install for public or non-sensitive social content work. Do not provide confidential launches, private campaign strategy, customer names, or unreleased business details unless you are comfortable with those details being used in web search queries; ask the agent to generalize or skip research when needed.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
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: 95% confidence
Finding: The skill directs the agent to use WebSearch with the user's niche or topic before drafting content, which can transmit potentially sensitive business context, campaign plans, or unpublished topics to external services without user awareness. This is primarily a privacy and data-handling issue rather than code execution, but it is still a real vulnerability because the workflow makes external disclosure the default and provides no consent or minimization guidance.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal