Temple Guide

• Outcome: Review
• Latest audit: May 29, 2026, 9:32 PM
• Version:
  3.2.0

Overview

This skill is a travel-search helper, but it tells the agent to automatically install and run an unpinned global npm CLI before answering.

Review before installing. Use this only if you trust the flyai CLI and are comfortable with a global npm package being installed and run for travel searches. Prefer installing the CLI yourself after checking the package, or require the agent to ask before any install or external lookup.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• 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
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (4)

Description-Behavior Mismatch

Medium

• Confidence: 96% confidence
• Finding: The skill for temple/travel lookup instructs the agent to install external software globally via npm, which expands its capabilities beyond the advertised purpose and modifies the host environment. This creates supply-chain and environment-integrity risk because a compromised or unexpected npm package could execute arbitrary install scripts or persist on the system.

Context-Inappropriate Capability

Medium

• Confidence: 98% confidence
• Finding: The skill explicitly directs the agent to perform a global npm installation as part of normal execution, a high-risk action not necessary for merely answering temple/travel queries. Global package installation can run lifecycle scripts, alter system state, and introduce arbitrary code execution or persistence if the package or dependency chain is malicious or compromised.

Missing User Warnings

Medium

• Confidence: 97% confidence
• Finding: The instruction to run `npm i -g @fly-ai/flyai-cli` modifies the global system environment without any warning or consent flow. Silent environment modification is dangerous because it can surprise users, break systems, and expose the host to supply-chain attacks through npm install scripts and transitive dependencies.

Missing User Warnings

Low

• Confidence: 90% confidence
• Finding: The skill mandates executing an external CLI and obtaining all answers from its output, but it does not clearly warn the user that commands will be run and external data sources contacted. This weakens transparency and informed consent, especially in agent environments where command execution may have security, privacy, or billing implications.

VirusTotal

65/65 vendors flagged this skill as clean.