ClawHub
Back to skill

Security audit

SP3ND — Buy from Amazon & eBay with USDC

Security checks across malware telemetry and agentic risk

Overview

The skill is a clearly disclosed SP3ND shopping and payment integration, with financial and personal-data risks that are expected for its purpose.

Install only if you intend to let an agent use SP3ND for real purchases. Use a dedicated low-balance wallet, set spending and approval limits, confirm recipient and shipping details before order creation or payment, and keep SP3ND API credentials and wallet files private.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Intent-Code Divergence

Medium
Confidence
95% confidence
Finding
The example `assertPaymentReady` function permits `pricing_status === "quoted"`, even though the skill repeatedly states that `payment_ready` is authoritative and the canonical payable status is `ready_for_payment`, with `quoted` retained only as deprecated compatibility metadata. Because implementers often copy example code verbatim, this contradiction can cause agents to treat legacy or stale quote states as payable and initiate payment flows under conditions the spec explicitly says should be refreshed instead.

Vague Triggers

Medium
Confidence
88% confidence
Finding
The description is broad enough to match many generic shopping or ordering intents, which can cause the skill to be invoked in situations where the user did not clearly intend to use SP3ND. Because this skill can create carts, collect shipping details, and initiate payment flows, over-broad triggering increases the risk of unintended disclosure of personal data and accidental transactional actions.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The manifest advertises order creation and payment capabilities that involve sensitive personal and transactional data, including shipping address, phone number, email, wallet attribution, and payment initiation, but it does not include an explicit user warning or consent-oriented usage boundary. In an agent ecosystem, that omission can make it easier for an orchestrator or model to route sensitive purchase flows without sufficiently informing the user that personal data will be transmitted to a third-party merchant service.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.