ClawHub
Back to skill

Security audit

explore-cambodia

Security checks for vulnerabilities and agentic risk

Overview

This Cambodia flight-search skill is mainly travel-related, but it should be reviewed because it can make an agent install a global FlyAI CLI package and has confusing command-scope rules.

Install only if you trust the FlyAI npm package and are comfortable sharing flight-search details with FlyAI/Fliggy. Require approval before any global npm install, prefer a pinned or sandboxed installation, and avoid relying on the broad 'discover' trigger unless you intend to invoke this Cambodia travel skill.

SkillSpector

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 (2)

Intent-Code Divergence

Medium
Confidence
94% confidence
Finding
The skill explicitly says agents must never invent CLI parameters, yet later instructs use of `--journey-type 1`, which is not listed in the allowed Parameters table. This inconsistency can cause agents to execute unsupported commands or trust undocumented behavior, weakening command safety guarantees and making future malicious flag smuggling easier to hide.

Missing User Warnings

Medium
Confidence
98% confidence
Finding
The skill mandates checking for a CLI and then globally installing `@fly-ai/flyai-cli` with `npm i -g` if missing, without explicit user consent or sandboxing. That directs the agent to modify the host environment and execute newly installed code from an external package, creating supply-chain, persistence, and system-integrity risk.

Static analysis

No suspicious patterns detected.