Shadows Skill Factory

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:16 PM UTC
• Version:
  1.1.0

Overview

This is a coherent text-only helper for creating new skill instruction files, with normal caution needed because generated skills can affect future agent behavior.

Safe to install as a skill-authoring helper, but review every generated SKILL.md before enabling, testing, or publishing it. Pay particular attention to broad triggers, shell commands, file or network access, credential requirements, and whether high-impact actions require explicit user approval.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (2)

Intent-Code Divergence

Medium

• Confidence: 96% confidence
• Finding: The skill's SECURITY CONSIDERATIONS state it does not modify files, but the documented process and output explicitly direct creating a directory and writing a new `SKILL.md` file. This mismatch can mislead users and downstream agents about the skill's side effects, weakening trust boundaries and causing file writes to occur without appropriate scrutiny or consent.

Vague Triggers

Medium

• Confidence: 84% confidence
• Finding: The trigger phrases include broad everyday language such as 'create a skill' and 'make a skill', which can cause the meta-skill to activate in unintended contexts. Because this skill generates agent instructions and files, accidental activation could lead to undesired workflow changes, prompt injection propagation into generated skills, or unsolicited file creation.

VirusTotal

66/66 vendors flagged this skill as clean.