Map Search

Security checks across malware telemetry and agentic risk

Overview

This is a user-run map search tool that sends searches and optional nearby-location data to map providers, with no hidden destructive or credential-harvesting behavior found.

Install only if you are comfortable sharing map searches and nearby-search location data with Amap, Baidu, or Tencent. Use map-only API keys with quotas, avoid sensitive queries, prefer explicit coordinates over automatic IP location when privacy matters, and treat the URL-encoding issue as something the publisher should fix.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (7)

Intent-Code Divergence

Medium

Confidence: 84% confidence
Finding: The documentation says provider priority will stop after the first valid result, but the code actually queries every configured provider and aggregates the responses. This mismatch can mislead users about how much data is shared externally and may cause unnecessary transmission of search terms to multiple third parties.

Intent-Code Divergence

High

Confidence: 95% confidence
Finding: The skill advertises nearby search and automatic IP-based geolocation, but the provided code does not implement those features. Security-relevant documentation drift is dangerous because users may believe the tool handles location data in a specific way, or may install and trust functionality that is not actually present, masking future code changes that add location collection without proper review.

Description-Behavior Mismatch

Medium

Confidence: 96% confidence
Finding: The code automatically determines the user's approximate location via an external IP geolocation request when coordinates are omitted, even though the skill presents itself as a map search tool rather than a location-discovery tool. This creates an unexpected privacy-sensitive behavior and can disclose location-derived data to third-party providers without explicit user intent or informed consent.

Intent-Code Divergence

Medium

Confidence: 93% confidence
Finding: The module documentation states that nearby search requires latitude/longitude, but the implementation silently infers location if they are missing. This documentation/behavior mismatch is security-relevant because users and calling agents may make privacy decisions based on the documented interface and not expect implicit geolocation or external transmission of location data.

Missing User Warnings

Medium

Confidence: 91% confidence
Finding: The skill description does not warn that user-entered search terms and region/location information are transmitted to external providers such as Amap, Baidu, and Tencent. This omission can expose sensitive queries or location context without informed user consent, especially in privacy-sensitive deployments.

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: Advertising automatic IP-based location detection without a prominent warning creates a privacy risk because users may unknowingly disclose approximate location to an external service. Location data is particularly sensitive, and implicit collection can violate user expectations and platform privacy requirements.

Missing User Warnings

Medium

Confidence: 98% confidence
Finding: When nearby search runs without provided coordinates, the skill auto-detects location and then sends that location to third-party map providers. Because there is no meaningful warning, confirmation, or consent gate before transmitting sensitive location information externally, this is a genuine privacy vulnerability.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal