Nm Abstract Modular Skills

• Outcome: Pass
• Latest audit: Jun 7, 2026 at 9:43 PM UTC
• Version:
  1.8.6

Overview

This is a documentation-only skill for designing modular skills; it has some overbroad activation and a broad chmod troubleshooting example, but no hidden execution, credential access, persistence, or exfiltration behavior.

Safe to install if you want modular skill-design guidance. Be aware it may activate on generic architecture or token discussions, and review any shell commands before running them, especially the recursive chmod example; prefer changing execute permissions only on specific trusted scripts.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Behavioral AST: exec() Call, eval() Call, Dynamic Import
• 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)

Vague Triggers

Medium

• Confidence: 90% confidence
• Finding: The trigger list includes broad generic terms such as 'architecture', 'modularity', 'tokens', and 'skills', which can cause the skill to activate in unrelated conversations. Over-broad activation increases the chance that the agent loads unnecessary instructions or follows this skill in contexts where it was not intended, which can interfere with higher-priority task handling and widen the attack surface for instruction confusion.

Missing User Warnings

Medium

• Confidence: 88% confidence
• Finding: The troubleshooting guidance includes a recursive bulk permission change (`find scripts/ -type f -exec chmod +x {} \;`) that will mark every file under `scripts/` as executable without narrowing scope or warning the user about broad filesystem changes. In a troubleshooting context, users may copy-paste commands directly; this can unintentionally alter many files, mask permission hygiene problems, and make non-executable content runnable if later invoked by other tooling.

VirusTotal

64/64 vendors flagged this skill as clean.