ClawHub
Back to skill

Security audit

极鲸云 Shein 评论搜索&分析

Security checks across malware telemetry and agentic risk

Overview

The skill’s review-search function is mostly coherent, but it reuses and stores login tokens under an unrelated Temu-named auth namespace, which creates a real credential-scoping concern.

Review this before installing if you use multiple GeekBI skills or accounts. The skill appears intended to query SHEIN reviews, but its login state may be shared through a Temu-named local auth file; install only if you are comfortable with that credential-sharing behavior or the publisher fixes the auth namespace.

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
92% confidence
Finding
The skill instructs the agent to read local reference files, invoke local Python scripts, and potentially access external services, but it does not declare corresponding permissions. Undeclared capabilities weaken reviewability and consent boundaries because the runtime behavior includes file access, shell execution, and network use that are not surfaced in metadata.

Description-Behavior Mismatch

Medium
Confidence
95% confidence
Finding
The auth state is stored under a different namespace (`temu-research-skill`) than the declared SHEIN review skill. This can cause credential/state confusion across skills, leading to accidental token sharing, incorrect account context, or one skill reading another skill's persisted auth state if they run under the same user environment. In an auth helper, cross-skill state mix-ups are a real security boundary failure, not just a naming bug.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.