ClawHub

微信跨境支付接入Skill

Security checks across malware telemetry and agentic risk

Overview

This is a documentation-only WeChat Pay integration skill, but it needs review because some payment-compliance guidance is inconsistent or under-scoped for automatic debits.

Review this carefully before using it for a real payment integration. It appears non-executable and not malicious, but users should verify direct-connection eligibility and automatic-debit notification requirements against official WeChat Pay documentation and local legal/compliance requirements before relying on the guidance.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Intent-Code Divergence

Medium
Confidence
96% confidence
Finding
The document states in the preface that only Hong Kong permits direct merchant integration, but later sections say both Hong Kong and the UK support direct connection. In a payments/compliance integration skill, contradictory eligibility rules can mislead implementers into choosing a non-compliant onboarding path, causing failed integrations, contractual issues, or regulatory handling mistakes.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The document explicitly describes delegated/automatic debit as '用户无感知（也可短信/邮件通知）', which frames customer notification as optional at the point of charging. In a payments integration skill, this can lead implementers to design silent recurring charges, increasing the risk of unauthorized debits, regulatory non-compliance, disputes, and fraud-like abuse even though later sections recommend notification.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal