Back to skill

Security audit

VIA Commerce

Security checks for vulnerabilities and agentic risk

Overview

This commerce skill is coherent, but it asks agents to handle payments, connect to third-party seller or buyer MCP endpoints, and collect a dashboard password without enough scoping or credential-safety guidance.

Review this skill carefully before installing. It appears to do what it claims, but it can route shopping and selling requests to an external commerce network, connect to participant-provided MCP servers, collect wallet and account-registration details, and initiate payment workflows after approval. Users should avoid reusing passwords, confirm endpoint and seller identity out of band for important purchases, and set strict payment delegation caps.

Vulnerability Patterns
  • Tool Hijacking and SpoofingModifies or replaces tools so legitimate-looking calls execute attacker logic
  • 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
Findings (1)

T07 · Tool Hijacking and Spoofing

Warning
Location
SKILL.md:41
Finding

Untrusted Dynamic MCP Endpoint Connection

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The skill declares very broad activation cues such as 'buy X', 'source X', and 'what's available on VIA', which can cause the agent to invoke an external commerce workflow for ordinary shopping-related queries without sufficiently clear user intent. In this context, unintended activation is meaningful because the skill is connected to a public MCP service and can progress from discovery toward seller interaction and eventual payment flows, increasing the chance of unnecessary data sharing or risky transactional actions.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The store registration workflow instructs the agent to collect sensitive data including a payout wallet, contact email, and notably a dashboard password, but provides no privacy notice, minimization guidance, or safe credential-handling constraints. This is dangerous because an agent may solicit, store, transmit, or log credentials and financial identifiers through the MCP flow without clear protections, creating risk of credential exposure, account takeover, and mishandling of wallet-linked identity information.

Content

No source excerpt is available for this finding.