Master Content Writer

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:38 PM UTC
• Version:
  1.0.0

Overview

This is a non-executable writing template for Chinese social-media drafts, with only minor scope and metadata notes.

Install only if you want help drafting Chinese social-media content. Be aware that ambiguous writing prompts may produce platform-style Chinese drafts, and verify the package identity because the metadata slug/version does not perfectly match the SKILL.md.

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: 84% confidence
• Finding: The activation triggers include generic phrases like '写文章', 'generate content', and 'create article', which can match many ordinary writing requests unrelated to Chinese social-media drafting. In agent environments, overly broad auto-activation can cause the skill to intercept prompts outside its intended scope, leading to incorrect behavior, unwanted steering toward platform-specific output, and possible policy bypass through tool misrouting.

Natural-Language Policy Violations

Medium

• Confidence: 76% confidence
• Finding: The skill description hard-codes Chinese social-media output without stating that language and locale should be confirmed with the user. This can cause the agent to override user expectations or route content generation into a specific language/market context by default, which is a scoping and consent issue, though not typically a severe security flaw by itself.

VirusTotal

64/64 vendors flagged this skill as clean.