ClawHub
Back to skill

Security audit

Travel Slow Experience Architect

Security checks across malware telemetry and agentic risk

Overview

This is a low-risk travel-planning skill that only formats user-provided trip details into descriptive slow-travel recommendations.

Safe to install for slow-travel planning. Users should still verify destinations, prices, safety advisories, and cultural details with current official sources because the skill does not use live data.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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)

Vague Triggers

Medium
Confidence
91% confidence
Finding
The trigger conditions are broad and ambiguous, including phrases like 'experience', 'pace if applicable', and generic travel-planning language. This can cause accidental invocation in unrelated user conversations, leading to prompt-routing confusion, unintended disclosure of user context to the wrong skill, or policy bypass through overly permissive activation.

Vague Triggers

Medium
Confidence
89% confidence
Finding
The example prompts use generic phrases such as 'help me', 'provide framework', and 'create checklist' that overlap heavily with normal travel-assistance requests. In a multi-skill environment, this increases the likelihood of over-broad matching and accidental invocation, which can degrade security boundaries between skills and produce unintended behavior.

VirusTotal

60/60 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.