ClawHub
Back to skill

Security audit

border-buddy

Security checks for vulnerabilities and agentic risk

Overview

This skill is an offline travel-rule helper with no hidden access, persistence, credential use, or data exfiltration behavior found.

Before relying on this for travel decisions, treat it as a first-pass offline snapshot and verify current rules with the named official authority. For Schengen calculations, use departure dates as exclusive exit dates and keep a safety margin because border authorities and official calculators may differ on edge cases.

SkillSpector

Vulnerability Patterns
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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 (1)

Intent-Code Divergence

Medium
Confidence
89% confidence
Finding
The document states that the exit day does not count, but the example input and surrounding examples are inconsistent enough to create ambiguity about whether `exit` is inclusive or exclusive. In a travel-compliance skill, that off-by-one ambiguity can cause users to miscalculate lawful stay time and potentially overstay Schengen limits, leading to denied entry, fines, or immigration consequences.

Static analysis

Detected: suspicious.dynamic_code_execution

Dynamic code execution detected.

Critical
Code
suspicious.dynamic_code_execution
Location
scripts/test_border_buddy.py:14