Phy Xiaohongshu Gtm

• Outcome: Pass
• Latest audit: May 28, 2026 at 6:38 AM UTC
• Version:
  1.0.2

Overview

This appears to be a Xiaohongshu marketing guidance skill, with minor usability concerns but no evidence of hidden execution, data theft, or destructive behavior.

Safe to install for Xiaohongshu marketing and content planning. Be aware that it may activate broadly on Xiaohongshu topics and may prefer Chinese output; users who need another language should explicitly request it.

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: 94% confidence
• Finding: The trigger description uses broad activation phrases such as "any 小红书 planning," which can cause the skill to activate outside narrowly intended contexts. Overbroad triggering increases prompt-injection surface and can hijack unrelated conversations, leading the agent to apply this skill when the user did not request Chinese Xiaohongshu marketing guidance.

Natural-Language Policy Violations

Medium

• Confidence: 90% confidence
• Finding: The skill mandates Chinese-only output regardless of user preference, which overrides user intent and can degrade safety, transparency, and usability in multilingual contexts. A rigid language requirement may also obscure model behavior from users or reviewers who do not read Chinese, making misaligned or harmful outputs harder to detect and contest.

VirusTotal

No VirusTotal findings