ClawHub
Back to skill

Security audit

Flyai Super Train / 火车票智能中转

Security checks for vulnerabilities and agentic risk

Overview

This train-ticket helper is mostly coherent, but it keeps detailed local travel and preference history without enough user control or retention guidance.

Install only if you are comfortable with the skill storing local train-trip history and preferences. Before using it for sensitive travel, consider clearing assets/history.json, avoiding saved defaults, and reviewing the external flyai CLI and API-key provider you will be sending searches to.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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 (4)

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill explicitly reads and writes persistent preference and history files containing travel history, seat choices, and purchasing outcomes, but does not require clear upfront consent or provide a meaningful privacy notice at the point of collection. In a travel context, this data can reveal movement patterns and habits, making silent persistence materially privacy-sensitive.

Ssd 3

Medium
Confidence
97% confidence
Finding
The skill instructs persistent storage of detailed user activity, including origin, destination, departure date/time, transfer city, full transport numbers, seat choices, and whether the user purchased or abandoned a trip. This creates a durable behavioral log that can expose sensitive travel patterns and decisions beyond the immediate task, especially if local files are accessible to other tools or users.

Ssd 3

Medium
Confidence
95% confidence
Finding
The workflow encourages long-term learning and reuse of user preferences from persistent files, which extends data use beyond the immediate request without strong consent safeguards. In this context, preference persistence can reveal recurring routes, accessibility needs, and personal travel habits, increasing privacy risk over time.

Ssd 3

Medium
Confidence
98% confidence
Finding
Automatically converting strongly worded user statements into persistent hard constraints without confirmation causes the assistant to retain long-term behavioral data based on a single utterance. This is risky because users may express temporary preferences or sensitive needs in natural language, and the skill turns them into stored profile data without explicit consent.

Static analysis

No suspicious patterns detected.