ClawHub
Back to skill

Security audit

Edvisage B2a Commerce

Security checks for vulnerabilities and agentic risk

Overview

This is a documentation-only commerce safety skill that clearly discloses payment-related behavior and does not include executable code, hidden install steps, or persistence mechanisms.

Before installing, treat this as payment-safety guidance rather than a complete payment implementation. If enabling real wallet transactions, confirm the spending-limit logic you use is internally consistent, keep the wallet minimally funded, use an approved-service whitelist, and verify logs do not expose private keys or sensitive wallet material.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

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
Category
Not specified by scanner
Confidence
95% confidence
Finding

The documented defaults create a contradictory control scheme: any transaction above $0.50 is already forbidden by the per-transaction limit, so the $1.00 human-authorization threshold can never be reached during normal operation. This can mislead users into believing there is an approval escalation path for larger payments when, in practice, the skill only hard-blocks them, which weakens policy clarity and can cause unsafe downstream implementations if integrators adjust one limit without understanding the dependency.

Content

No source excerpt is available for this finding.