Ai Content Detection

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:21 PM UTC
• Version:
  0.1.0

Overview

This is a guidance-only AI content detection skill with no code, install hooks, persistence, or hidden actions.

Installers should treat this as an informational forensic checklist. Before using third-party detectors or forensic services, consider whether private documents, images, audio, video, URLs, or analytics data would be exposed, and treat any AI-detection result as uncertain rather than definitive.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• 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 skill activation text is extremely broad and includes many loosely related triggers such as authenticity checking, online AI-content statistics, false accusation defense, contracts, articles, links, and multiple media types. In an agentic system, this can cause the skill to be invoked for common trust/safety or content-analysis requests beyond its intended scope, increasing the chance of inappropriate tool use, overconfident forensic claims, or misrouting sensitive requests to a skill with unreliable detection guidance.

Natural-Language Policy Violations

Medium

• Confidence: 82% confidence
• Finding: The skill content is written as a Chinese-only operating guide and does not offer a language-selection step despite metadata claiming relevance regardless of language. If the agent follows the skill literally, users may receive analysis or structured reports in an unexpected language, causing misunderstanding, poor consent, and higher risk of miscommunication in sensitive authenticity or fraud assessments.

VirusTotal

65/65 vendors flagged this skill as clean.