ClawHub
Back to skill

Security audit

Book Cleaning

Security checks for vulnerabilities and agentic risk

Overview

This skill is a coherent cleaning-booking integration, but it can send personal contact details to an external booking service and create bookings without clear consent or confirmation guidance.

Review this skill before installing. It appears purpose-aligned and not malicious, but users should only use it when they intentionally want Lokuli involved, and the agent should confirm the selected provider, time, and personal contact details before sending them or creating a booking.

A.I.G

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

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Missing User Warnings

High
Confidence
98% confidence
Finding
The skill facilitates booking through an external MCP endpoint and the booking example includes customer name, email, and phone number, but the description does not warn that this personal data will be sent to a third-party service. This creates a privacy and consent risk because users may not understand that their PII is being transmitted off-platform during booking.

Vague Triggers

Medium
Confidence
95% confidence
Finding
The trigger text is broad enough to activate on almost any cleaning-related request, including vague phrases like 'any cleaning service request.' Overbroad activation can cause the agent to invoke an external booking workflow without sufficiently clear user intent, increasing the chance of unintended third-party calls or premature collection/transmission of personal data.

Static analysis

No suspicious patterns detected.