ClawHub
Back to skill

Security audit

Agent身份助手专业版

Security checks across malware telemetry and agentic risk

Overview

The skill is broadly aimed at enterprise agent-persona management, but it asks for powerful read/write/exec abilities and includes under-scoped credential, callback, import/export, and agent-behavior integration guidance.

Review this skill before installing. It is not clearly malicious, but it should be used only when you intentionally want an agent to manage persona files, run related commands, migrate or export data, and integrate with external systems. Prefer environment variables or a secret manager for tokens, disable callbacks unless needed, and require explicit approval before writes, migrations, exports, routing changes, or command execution.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Intent-Code Divergence

Medium
Confidence
94% confidence
Finding
The skill claims secrets should be provided securely, yet instructs storing a SkillHub token in a local plaintext file path. This creates a real secret-handling weakness because local files are easier to leak through backups, misconfigured permissions, malware, or accidental reads by other tools with filesystem access.

Vague Triggers

High
Confidence
88% confidence
Finding
The trigger condition is so broad that it can activate for almost any AI, agent, orchestration, or LLM-related request. In this skill's context, broad activation is dangerous because the skill exposes read/write/exec capabilities and could be selected in situations where users did not intend file changes, command execution, routing, exports, or external integrations.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill advertises file operations, command execution, exports/imports, and callback URLs, but does not prominently warn about data modification, migration side effects, or outbound transmission risks near those features. That omission can mislead users into invoking destructive or privacy-impacting actions without understanding that data may be changed, exported, or sent to external systems.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.