ClawHub
Back to skill

Security audit

mayu-cn

Security checks across malware telemetry and agentic risk

Overview

This skill does not show malware-like behavior, but its always-on global prompt behavior is broad enough that users should review it before installing.

Install only if you intentionally want this skill to influence agent behavior broadly across conversations. Prefer an explicit opt-in trigger or disabling always-on/global activation if you want it limited to deliberate use.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Vague Triggers

High
Confidence
95% confidence
Finding
The skill is configured as global and always-on, while its trigger set includes broad natural-language terms such as 'AI行为', '自省', and '边界' that can appear in ordinary conversations. This creates a prompt-surface hijacking risk where the skill activates unintentionally and injects a large behavioral prompt into unrelated sessions, potentially overriding user intent, altering responses, or increasing susceptibility to prompt-conflict and instruction persistence.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.