智能多面手工具箱

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:35 PM UTC
• Version:
  1.0.0

Overview

This appears to be a simple local utility skill, with broad invocation wording but no evidence of hidden access, persistence, or data exfiltration.

Reasonable to install if you want a local general-purpose helper. Invoke it intentionally because its trigger style is broad, and avoid passing secrets or private content unless you are comfortable with the agent processing that input.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• 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
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (1)

Vague Triggers

High

• Confidence: 96% confidence
• Finding: The skill advertises very broad natural-language invocation such as handling loosely phrased requests across querying, ideation, tooling, and quality review. In an agent ecosystem, this can cause accidental activation during ordinary conversation, leading the skill to intercept prompts outside its intended scope, potentially influencing routing, exposing data to unnecessary processing, or bypassing user expectations about which capability is being used.

VirusTotal

67/67 vendors flagged this skill as clean.