ClawHub
Back to skill

Security audit

boutique-hotel

Security checks across malware telemetry and agentic risk

Overview

This travel-search skill is mostly purpose-aligned, but it should be reviewed because it directs agents to make an automatic global npm install and contains an inconsistent command example.

Review before installing. If you use it, install or verify the flyai CLI yourself from a trusted source, avoid letting an agent run the global npm install automatically, and treat booking links as external commercial links. Also confirm whether --journey-type is actually supported by the current CLI before relying on direct-flight results.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Intent-Code Divergence

Medium
Confidence
92% confidence
Finding
The skill explicitly says agents must never invent CLI parameters, yet the direct-flight playbook uses `--journey-type 1`, which is absent from the documented parameter table. In an agent setting, this kind of contradiction can cause the model to execute undocumented commands or normalize unsupported flags, undermining command safety guarantees and making behavior less auditable.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The skill instructs the agent to run `npm i -g @fly-ai/flyai-cli` automatically when the tool is missing, which modifies the host environment without explicit user approval. In a security context, silent global package installation is dangerous because it expands system attack surface, can introduce supply-chain risk, and crosses a trust boundary from answering queries into changing the machine.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.