Content Rewriter 内容多平台改写

• Outcome: Pass
• Latest audit: May 29, 2026, 8:14 PM
• Version:
  1.0.0

Overview

This is a prompt-only skill for drafting Chinese social media content, with no evidence of code execution, credential use, persistence, or hidden external actions.

Safe to install for Chinese social-media drafting. Make sure it is invoked only when you want Chinese platform-specific content, and review generated claims, citations, CTAs, and platform compliance before publishing manually.

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: 82% confidence
• Finding: The activation trigger list includes very broad generic terms like 'rewrite' and '转载', which can match normal user requests outside the intended narrow use case. In agent environments with automatic skill routing, this can cause unintended invocation, leading the skill to rewrite content when the user did not explicitly request Chinese social-platform adaptation.

Natural-Language Policy Violations

Medium

• Confidence: 76% confidence
• Finding: The skill is explicitly designed to repurpose content for Chinese social-media platforms and applies Chinese platform conventions by default, but it does not require explicit user confirmation that this locale-specific transformation is desired. In a multi-skill agent setting, unintended activation could force outputs into the wrong language, market, or compliance context, causing misalignment with user intent and possible policy or publishing issues.

VirusTotal

66/66 vendors flagged this skill as clean.