Back to skill

Security audit

Hokkaido Travel Companion

Security checks for vulnerabilities and agentic risk

Overview

This Hokkaido travel skill is a disclosed local travel-planning workflow with proportionate local note and budget storage, but users should understand what personal trip details may be saved.

Before installing, expect this skill to save Hokkaido trip notes and budget-related details locally. Avoid entering hotel, itinerary, budget, or visit notes you would not want retained, and review or delete the local vault and budget files before sharing or publishing the skill directory.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (3)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The core loop directs automatic writes to a local vault after search and synthesis, including reusable travel knowledge and potentially user preferences or visit history, but does not tell the user that this information will be retained. This is dangerous because it normalizes silent persistence of personal or behavior-linked data, increasing the chance of privacy surprises, over-collection, and unintended later disclosure.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The transport output format requires every route to include station names in Japanese, Korean, and English, and the vault node format also includes Korean text placeholders. This imposes a specific locale/language requirement without user opt-in, which can violate language-choice policy.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill instructs the agent to collect and persist personal travel details such as budget, trip dates, hotel-nearest station, and IC card type in a local file without an explicit consent, retention, or privacy notice. While this is not code execution or exfiltration, it creates a real privacy risk because users may not realize their itinerary and location-related data is being stored for later reuse.

Content

No source excerpt is available for this finding.