ClawHub
Back to skill

Security audit

Unique Homestay

Security checks for vulnerabilities and agentic risk

Overview

This travel-planning skill mostly matches its purpose, but it appears to keep hidden local logs of users’ travel requests, which can expose personal trip details.

Review before installing. The skill is not clearly malicious, but use it only if you are comfortable installing a third-party flyai CLI and potentially having travel prompts stored locally. Avoid entering passport, visa, payment, or other sensitive personal details unless the logging behavior is removed or clearly controlled.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • System Prompt LeakageDirect Leakage, Indirect Extraction, Tool-Based Exfiltration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The runbook explicitly records the raw `user_query` and appends the generated log to a local file, while also stating the log is not shown to users. This creates undisclosed retention of potentially sensitive travel data, personal identifiers, or booking details, increasing privacy and data-handling risk if the file is later accessed, copied, or exfiltrated.

Ssd 3

Medium
Confidence
94% confidence
Finding
The runbook instructs the agent to retain raw input in an internal execution log and persist it when filesystem writes are available. Because this skill operates in a travel-booking context where users may provide names, dates, destinations, visa details, and other personal information, retaining verbatim prompts materially increases exposure of sensitive data without any stated need-to-know limitation or user disclosure.

Direct Prompt Extraction

High
Category
System Prompt Leakage
Content
flyai search-hotel --dest-name "Dali" --hotel-types "民宿" --sort rate_desc --check-in-date 2026-05-01 --check-out-date 2026-05-03
```

## Output Rules

1. **Conclusion first** — lead with the key finding
2. **Comparison table** with ≥ 3 results when available
Confidence
87% confidence
Finding
The skill contains strong instruction blocks that attempt to control agent behavior, including mandatory CLI execution, installation steps, strict output gating, and references to external files for operational guidance. In a hostile or untrusted skill ecosystem, this is dangerous because an agent may prioritize these embedded instructions and perform unnecessary command execution or package installation based solely on the skill content, increasing supply-chain and prompt-injection risk.

Static analysis

No suspicious patterns detected.