Senior Frontend

Security checks across malware telemetry and agentic risk

Overview

This is a markdown-only frontend design skill with broad activation language but no hidden code, credential use, persistence, or data exfiltration behavior.

Installers should expect this skill to activate on many frontend UI requests and to favor the Next.js/Tailwind/shadcn stack. Confirm the framework you want when using it, review any generated dependency commands before running them, and treat the unrelated crypto/purchase metadata tags as something the publisher should correct.

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 (2)

Vague Triggers

High

Confidence: 94% confidence
Finding: The skill metadata includes very broad auto-trigger phrases such as 'build me a page', 'create a component', and 'review my UI', which can match ordinary user requests and cause the skill to activate without clear user intent. This is dangerous because it can unexpectedly override other skills or force a frontend-specific workflow and technology stack onto unrelated tasks, reducing user control and increasing the chance of prompt-routing mistakes.

Natural-Language Policy Violations

Medium

Confidence: 87% confidence
Finding: The skill states that output is always 'production-ready Next.js + Tailwind CSS + shadcn/ui' code, regardless of whether the user requested that stack. This is risky because it constrains agent behavior without consent, potentially producing unsuitable or incompatible deliverables and overriding user or system preferences.

VirusTotal

67/67 vendors flagged this skill as clean.

View on VirusTotal