YES.md 中文版

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:18 PM UTC
• Version:
  1.1.0

Overview

This is a Chinese instruction-only engineering workflow skill that is broad but transparent and does not include hidden code, persistence, credential handling, or data exfiltration.

Install this if you want a Chinese workflow skill that makes the agent gather evidence, use tools before guessing, back up files before risky edits, and verify results. Keep normal approval controls for production deployments, database changes, and commands that read sensitive environment details.

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: 94% confidence
• Finding: The activation criteria are so broad that this skill can trigger for routine development, debugging, deployment, configuration, and data tasks, effectively making it a near-global behavioral override. That increases the chance of unintended invocation, instruction conflicts with other skills, and over-application of rigid procedures in contexts where they may be unsafe, inapplicable, or disruptive.

VirusTotal

64/64 vendors flagged this skill as clean.