ClawHub
Back to skill

Security audit

Apple Maps (MacOS)

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly coherent for opening Apple Maps, but its command examples can invite unsafe shell interpolation of user-provided map queries.

Install only if you are comfortable with a macOS automation skill that opens Apple Maps and stores limited local preferences. Before use, ensure map query values are percent-encoded and launched as separate process arguments rather than interpolated into a shell command.

A.I.G

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (1)

T09 · Insecure Skill Coding Practices

Error
Location
command-paths.md:19
Finding

Command Injection Through Unsafe Interpolation of Map Parameters

Content
View full analysis
Remediation
View remediation

SkillSpector

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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)

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 91)May include surrounding context.

md
- Sharing raw links that include private notes -> accidental disclosure.
- Using UI scripting as default path -> brittle behavior across locales.
- Opening many candidate links at once -> user loses context quickly.
- Assuming route mode without confirmation -> wrong transport directions.

## External Endpoints