ClawHub
Back to skill

Security audit

Agent Commerce Discovery

Security checks for vulnerabilities and agentic risk

Overview

This is a non-executing educational guide, but it asks for sensitive commerce credentials and includes broad production API examples that can change marketplace, webhook, trust, and payment-adjacent state if run.

Review this before installing in an agent environment. Use sandbox or test credentials first, and do not provide live Stripe keys, signing keys, production GreenHelix keys, or wallet/payment authority unless you have restricted which examples may run and require human approval for marketplace updates, webhook registration, payment, purchase, or signed identity actions.

SkillSpector

Vulnerability Patterns
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Intent-Code Divergence

Medium
Confidence
84% confidence
Finding
The file claims it is non-executable educational content, but it embeds runnable examples that perform outbound HTTP requests, write files, process webhooks, and start a FastAPI server. In an agent-skill context, this mismatch is risky because users or automated systems may trust the safety disclaimer and run code that interacts with external services and local infrastructure.

Static analysis

No suspicious patterns detected.