Sorftime Product Search

• Outcome: Review
• Latest audit: May 29, 2026 at 9:33 PM UTC
• Version:
  1.0.0

Overview

This skill mostly does the advertised Amazon/Sorftime product search, but it also tells agents to automatically send free-form feedback to a separate external LinkFox endpoint without clear user consent.

Install only if you are comfortable sending Amazon product-search queries to LinkFox/Sorftime under your LinkFox API key. Treat the feedback feature carefully: do not allow automatic submission of personal, confidential, proprietary, or full conversation details unless you intentionally approve that disclosure.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands

Findings (4)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 87% confidence
• Finding: The skill instructs the agent to use an external LinkFox/Sorftime API and references direct script execution, which implies network and possibly environment-backed capabilities, yet no explicit permissions are declared. This creates a transparency and governance gap: a host system or reviewer may not realize the skill can transmit user-provided query data off-platform or rely on sensitive runtime configuration.

Context-Inappropriate Capability

Medium

• Confidence: 96% confidence
• Finding: The documented Feedback API introduces an additional outbound network capability unrelated to the stated product-search function. Because it can transmit free-form content to a separate external service, it expands the skill's data-flow surface and could be used to send user prompts, conversation summaries, or other sensitive context without being necessary for the core task.

Vague Triggers

Medium

• Confidence: 80% confidence
• Finding: The trigger conditions are intentionally expansive, activating not only for Sorftime-specific requests but for broad Amazon product research, filtering, comparison, and category/brand/seller exploration requests even when Sorftime is not mentioned. Overbroad activation can hijack unrelated user intents, cause unnecessary external API calls, and route user data to a third-party service without sufficiently specific user intent.

Missing User Warnings

Medium

• Confidence: 98% confidence
• Finding: The feedback documentation allows transmission of a `content` field containing user-provided text to an external service but does not require notice, consent, or data-minimization guidance. In an agent setting, this creates a realistic risk of exfiltrating user requests, internal reasoning summaries, or other sensitive information under the guise of feedback submission.

VirusTotal

67/67 vendors flagged this skill as clean.