ClawHub
Back to skill

Security audit

支付集成专家

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent payment-integration coding skill, with no hidden or destructive behavior found, though users should review generated payment and refund logic carefully.

Install only if you want an agent to help write or modify payment integration code. Use test API keys first, keep secrets in environment variables or a secret manager, confirm any exec/install command before running it, and manually review refund, subscription cancellation, webhook, and amount-validation logic before using generated code in production.

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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)

Intent-Code Divergence

Medium
Confidence
94% confidence
Finding
The skill contains contradictory guidance: one boundary condition says completed orders cannot be refunded, while multiple later sections explicitly describe refund handling for completed payments. In a payment integration skill, inconsistent business rules can cause implementers to deny legitimate refunds, mishandle post-payment logic, or build incorrect controls that create compliance, customer harm, and operational risk.

Vague Triggers

Medium
Confidence
87% confidence
Finding
The trigger keywords are broad terms such as payment integration, Stripe, webhook, automation, code, and integration, which are likely to match many ordinary conversations. In agent environments, overbroad activation can cause the skill to engage unexpectedly and gain access to read/write/exec capabilities in contexts where the user did not explicitly request this payment skill, increasing the attack surface and risk of unintended actions.

Static analysis

No suspicious patterns detected.