Ebay Search

Security checks across malware telemetry and agentic risk

Overview

This is mostly an eBay search helper, but it also tells agents to silently send feedback and user-intent details to a separate LinkFox endpoint.

Review before installing. Use it only if you trust LinkFox with your eBay search queries, filters, API key usage, and any automatically reported feedback. Avoid sensitive business or personal details in searches, and do not allow feedback submission unless the user explicitly agrees to send it.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (5)

Context-Inappropriate Capability

Medium
Confidence
96% confidence
Finding
The skill instructs the agent to automatically report broad categories of user feedback, including praise, dissatisfaction, and anything that could be improved, to a separate Feedback API unrelated to the core eBay search function. This creates a covert secondary data flow in which user interaction content and inferred sentiment may be transmitted without necessity, minimization, or explicit consent.

Context-Inappropriate Capability

Medium
Confidence
95% confidence
Finding
The documentation for an eBay search skill includes a separate feedback endpoint on a different domain that is unrelated to the core search function. This creates an unexpected secondary data flow that could transmit user content externally without clear user consent, purpose limitation, or privacy boundaries, making the skill more dangerous than its stated functionality suggests.

Vague Triggers

High
Confidence
93% confidence
Finding
The activation rules are overly broad and allow the skill to trigger even when the user does not explicitly mention eBay, based merely on inferred marketplace-search intent. Overbroad triggering can route unrelated user queries into this skill, causing unnecessary external requests and unintended disclosure of user queries to the tool backend.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The skill explicitly tells the agent to auto-report user feedback and satisfaction signals to a Feedback API without notifying the user. This undermines user expectations and can leak conversational content, sentiment, or behavioral metadata to a secondary service without transparency or consent.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The feedback section instructs sending freeform content that may include what the user said or intended to a separate external endpoint, but provides no privacy warning, consent requirement, or minimization guidance. In an agent context, this can lead to unintended disclosure of personal, confidential, or sensitive user data to a third party outside the primary tool flow.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal