ClawHub
Back to skill

Security audit

指边美妆登录会话 HTML 桥接

Security checks across malware telemetry and agentic risk

Overview

This looks like a real login bridge, but it handles SMS login/session data and stores session IDs locally without enough disclosure or protection.

Install only if you trust the publisher and intend to send SMS login details and HTML content to the documented zhibianai.com service. Treat CLAWHUB_SKILL_TOKEN and .session.json as sensitive secrets; delete or protect the session file after use, and prefer a revised version that asks before transmitting login data and stores session IDs with restrictive permissions or no persistence by default.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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)

Lp3

Medium
Category
MCP Least Privilege
Confidence
94% confidence
Finding
The skill documents access to environment variables, network endpoints, and local debug commands that imply file interaction, but it does not declare any permissions or capability boundaries. This is dangerous because an agent or reviewer cannot accurately assess what sensitive resources the skill may access, increasing the chance of unintended secret exposure or unauthorized local/network actions.

Tp4

High
Category
MCP Tool Poisoning
Confidence
92% confidence
Finding
The declared description says the skill is for login/session creation and HTML exchange, but the content also includes SMS sending, local session persistence behavior, and interactive multi-turn client workflows beyond that stated scope. Description-behavior mismatch is dangerous because it prevents informed consent and hides materially different operations, especially those involving external messaging and local storage of session identifiers.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill sends highly sensitive data—phone numbers, verification codes, session identifiers, and user-supplied HTML—to external endpoints without an explicit user-facing warning in the description. This is dangerous because users may disclose credentials and personal data without understanding they are being transmitted to a third-party service, creating privacy and account-takeover risk if mishandled.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal