Brand Guidelines

Security checks across malware telemetry and agentic risk

Overview

This is a text-only Anthropic brand styling guide with no code execution, data access, persistence, or hidden behavior.

Safe to install from a security perspective. Use it when you explicitly want Anthropic-style branding, and verify the brand colors, typography, and permission to use Anthropic branding before applying it to public or legal materials.

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 manifest description is broadly phrased to apply to 'any sort of artifact' and 'when brand colors or style guidelines, visual formatting, or company design standards apply,' which can cause over-invocation on generic design or formatting tasks. This is not malicious in context, but it increases the chance the skill is selected outside its narrow intended scope, potentially causing inappropriate post-processing or unintended modification of user outputs.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal