Security audit

极鲸云 Shein 图搜同款&图片搜索分析

Security checks across malware telemetry and agentic risk

Overview

The skill mostly matches its SHEIN image-search purpose, but it should be reviewed because it persists and reuses login state through a mismatched Temu-named auth path.

Install only if you are comfortable uploading product images to GeekBI and sharing GeekBI login state across related skills. Review the auth-state behavior first, especially if you also use other GeekBI or Temu-related skills on the same machine.

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: 91% confidence
Finding: The skill instructs the agent to read local files or attachments, invoke shell commands, and make network requests, but it does not declare corresponding permissions. This creates a capability/permission mismatch that can bypass review expectations and lead to unintended access to local data or external services if the skill is invoked with attacker-controlled inputs.

Description-Behavior Mismatch

Medium

Confidence: 88% confidence
Finding: The auth state path is hardcoded under "temu-research-skill" while this skill is a SHEIN image-search skill. That mismatch can cause credential/state confusion across skills, leading to unintended token reuse, cross-skill logout/login interference, or reading another skill's persisted auth state if both operate against related services on the same machine.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.