m

• Outcome: Review
• Latest audit: May 29, 2026, 9:24 PM
• Version:
  1.0.0

Overview

This is an instruction-only shortcut skill, but it bundles broad file, package, service, and data-migration commands under a one-letter trigger with insufficient safety boundaries.

Install only if you want a broad local command-reference skill and your agent setup requires explicit confirmation before running commands. Review every generated command before execution, especially deletion, rsync --delete, sudo package operations, service enable/startup changes, database restores, Docker volume copies, and Git tag operations.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (4)

Vague Triggers

High

• Confidence: 97% confidence
• Finding: Using a single-letter trigger like 'm' creates a high likelihood of accidental invocation during ordinary conversation or shorthand input. Because this skill includes file moves, package changes, and service management, unintended activation could lead to disruptive or destructive system actions.

Vague Triggers

Medium

• Confidence: 91% confidence
• Finding: The examples combine unrelated domains—file operations, package management, and service administration—so the activation boundary is unclear. This increases the chance that benign requests are interpreted as authorization for privileged or destructive actions outside the user's intended scope.

Missing User Warnings

Medium

• Confidence: 89% confidence
• Finding: The skill advertises package and service management alongside moving files without an upfront warning that these actions can alter system state, remove software, or interrupt running services. Users may invoke it expecting harmless file operations and instead receive guidance for impactful administrative changes.

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The database export/import examples handle sensitive data and can overwrite or expose production information, yet they omit warnings about credential handling, destination validation, and restoration risk. In an agent context, such examples can normalize unsafe handling of backups and live data migrations.

VirusTotal

64/64 vendors flagged this skill as clean.