查找资料

• Outcome: Review
• Latest audit: May 27, 2026, 10:14 PM
• Version:
  1.0.1

Overview

This search skill does what it advertises, but it exposes a shared API key and sends user searches to an unclear HTTP endpoint.

Review before installing. Use only non-sensitive searches unless the publisher replaces the embedded key with a user-managed credential, provides HTTPS on a clearly owned endpoint, and documents privacy, retention, caching, and cost controls.

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
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (4)

Context-Inappropriate Capability

High

• Confidence: 99% confidence
• Finding: The skill documentation directly embeds a usable Bearer API key, exposing credential material to any reader of the skill. This enables unauthorized use of the backend search service, potential quota theft, cost abuse, and use of the key outside the intended agent workflow.

Context-Inappropriate Capability

High

• Confidence: 99% confidence
• Finding: The same Bearer token is repeated in the news API section, increasing exposure and confirming that the credential is intended to be used directly by skill consumers. Repetition broadens leakage opportunities and makes unauthorized access to the third-party service trivial.

Intent-Code Divergence

Medium

• Confidence: 86% confidence
• Finding: The document claims the skill uses 'official APIs,' but the actual endpoint is a hard-coded IP address for an opaque aggregation service over plain HTTP. This misrepresentation can cause users and agents to trust a service they do not actually understand, while sending search queries and crawled content to an unverified intermediary.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The skill omits a user-facing warning that search queries and optionally crawled page content are transmitted to an external third-party API. In this context, users may submit sensitive prompts, proprietary research topics, or internal URLs, so the lack of disclosure meaningfully increases privacy and data-handling risk.

VirusTotal

VirusTotal findings are pending for this skill version.