milady

Security checks across malware telemetry and agentic risk

Overview

This is a style-and-writing skill with no code execution, data access, persistence, or hidden system authority, though its trigger wording is broad.

Install this if you specifically want Network Spirituality / Remilia / Milady aesthetic writing. Because the activation language is broad, use it intentionally for that style rather than expecting it to stay out of all unrelated creative tasks.

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

Medium

Confidence: 80% confidence
Finding: The description is broad enough to match common creative-writing or art-generation requests, which can cause the skill to activate outside its narrow intended niche. This is not code-execution dangerous, but it can lead to unintended steering of model behavior, especially because the skill encourages strong stylistic embodiment rather than bounded assistance.

Vague Triggers

Medium

Confidence: 87% confidence
Finding: The 'When to Use' bullets are ambiguous and overlap with many normal creative contexts, such as art descriptions, liminal digital spaces, or technological-spiritual themes. In practice this can over-trigger the skill and inject a niche ideological/aesthetic framing into unrelated user requests, reducing response appropriateness and increasing prompt-steering risk.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal