ClawHub
Back to skill

Security audit

Identity Emergence Field

Security checks across malware telemetry and agentic risk

Overview

This is a markdown-only reasoning framework for structuring complex tasks, with no evidence of hidden execution, data collection, persistence, or credential access.

Install this if you want a Chinese-language framework for decomposing complex tasks into roles, consensus steps, and report templates. Review the broad activation style and default-hidden reasoning detail if you prefer narrow, explicit skill invocation or full process transparency.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Vague Triggers

Medium
Confidence
91% confidence
Finding
The activation condition '任务需要多身份协作' is so broad that it can trigger the skill for a very wide range of tasks without clear boundaries, eligibility checks, or user confirmation. In an agent setting, overly broad activation criteria can cause unintended invocation, prompt-routing errors, and inappropriate application of hidden internal reasoning frameworks to tasks that do not need them.

Natural-Language Policy Violations

Medium
Confidence
85% confidence
Finding
The exemplar is entirely written in Chinese and presents outputs in Chinese without any indication of honoring user language preference. In a multi-user or multilingual agent environment, forcing a locale can degrade usability, miscommunicate safety-relevant content, and cause the agent to ignore explicit user preference or system locale expectations.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.