ZeeLin AI Detector

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:16 PM UTC
• Version:
  1.0.0

Overview

This is an instruction-only Chinese skill that estimates whether provided text sounds AI-written and does not request files, credentials, tools, network access, or persistent behavior.

Safe to install from the reviewed artifacts. Treat its percentage as a rough writing-style opinion, not proof of authorship or a professional AI-detection or plagiarism result; non-Chinese users may find the fixed Chinese format less convenient.

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 (2)

Vague Triggers

Medium

• Confidence: 90% confidence
• Finding: The trigger examples are broad, short, and conversational, which increases the chance this skill activates on ordinary user requests that may not explicitly opt into this specific detector workflow. Over-broad activation can cause unintended routing, surprise behavior, or interference with other more appropriate skills, especially because phrases like '检测AI痕迹' and 'AI查重' are common user wording.

Natural-Language Policy Violations

Medium

• Confidence: 84% confidence
• Finding: The skill metadata description is written only in Chinese and does not indicate language flexibility or user preference handling. In multi-language environments, this can force or bias routing and responses into a specific language without explicit user opt-in, causing confusion, degraded UX, or incorrect activation for users expecting another language.

VirusTotal

63/63 vendors flagged this skill as clean.