Map Route Reader

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:31 PM UTC
• Version:
  1.0.2

Overview

This instruction-only map route skill is purpose-aligned, but users should be mindful that route details and a selected map tab can reveal location information.

Install if you want an agent to query Amap or Baidu for route details. When Browser Relay is requested, share only the specific map route tab, avoid exposing saved places or unrelated account content, and verify time-sensitive or safety-critical travel details in the map app.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (4)

Vague Triggers

Medium

• Confidence: 91% confidence
• Finding: The manifest description is written to activate on a wide range of ordinary travel questions and lacks clear constraints on when the skill should not be used. In an agent environment, overly broad activation can cause the skill to intercept requests outside its intended scope, leading to incorrect tool routing, unnecessary browser use, or exposure of user browsing context when a simpler response would suffice.

Vague Triggers

Medium

• Confidence: 95% confidence
• Finding: The trigger list contains broad phrases like driving time, walking time, and route comparisons without qualification, which makes invocation boundaries ambiguous. This increases the chance the agent invokes the skill for generic planning questions, causing overreach into browser-assisted map extraction and potentially unnecessary access to user map tabs via Browser Relay.

Vague Triggers

Medium

• Confidence: 86% confidence
• Finding: The description uses very broad example phrasing such as any question about getting from A to B, comparing speed, convenience, taxi suitability, or alternatives, without defining precise invocation boundaries. That can cause over-triggering in unrelated conversations and increase the chance the skill is invoked when the user did not explicitly request live route retrieval, which may lead to unnecessary external-data use or incorrect tool selection.

Natural-Language Policy Violations

Medium

• Confidence: 80% confidence
• Finding: The description is written as a Chinese-only behavioral constraint and does not indicate that output language should follow user preference or locale. This can override normal language-selection behavior, causing mismatched responses, reduced user comprehension, and potentially unsafe misunderstandings when route instructions are delivered in a language the user did not request.

VirusTotal

64/64 vendors flagged this skill as clean.