ClawHub
Back to skill

Security audit

极鲸云 Shein 关键词搜索&分析

Security checks across malware telemetry and agentic risk

Overview

The skill appears to perform SHEIN keyword research as advertised, but its saved GeekBI login state is scoped in a confusing way that could share sessions across unrelated skills.

Review this before installing if you use multiple GeekBI/OpenClaw skills. It may reuse and mirror a GeekBI login token across local stores, including a TEMU-named user config location; install only if that shared-session behavior is acceptable, and use the provided clear command or remove the auth state if you want to reset access.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
86% confidence
Finding
The skill instructs the agent to read local reference files and invoke local Python scripts, which are code-capable actions, but it declares no permissions. This creates a governance and containment gap: reviewers and runtime policy may underestimate the skill's access to files, shell execution, and network-backed queries, increasing the risk of unintended data access or command execution if supporting scripts or inputs are compromised.

Description-Behavior Mismatch

Medium
Confidence
84% confidence
Finding
The auth state is stored under a shared-looking directory name, "temu-research-skill", even though this skill is for SHEIN keyword analysis. If another skill or tool uses the same path, authentication state can be unintentionally shared, overwritten, or cleared across unrelated skills, causing cross-skill token confusion and possible unauthorized reuse of a valid session. In an agent ecosystem where multiple skills coexist on the same host, this increases the chance of credential mix-ups.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.