ClawHub
Back to skill

Security audit

Book Appliance Repair

Security checks for vulnerabilities and agentic risk

Overview

This skill is a narrow appliance-repair booking helper that uses Lokuli's external MCP service and does not show hidden code, persistence, or deceptive behavior.

Before installing, understand that booking through this skill may send your name, email, phone number, preferred time, and repair request details to Lokuli's external MCP service. Use it when you intend to search or book appliance repair through Lokuli, and confirm the exact details before asking the agent to create 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
97% confidence
Finding
The skill includes a create_booking flow that sends customer name, email, and phone number to an external MCP endpoint, but the description does not warn the user that their personal contact information will be transmitted off-platform. This lack of transparency can cause unauthorized sharing of PII and makes accidental data disclosure more likely, especially when combined with the broad trigger behavior.

Vague Triggers

Medium
Confidence
97% confidence
Finding
The trigger language is broad enough to activate on generic 'service request' phrasing, which can cause the skill to run when the user did not clearly ask to book through this external provider. In a booking skill, over-triggering is risky because it can lead to unintended searches or reservation flows that collect or transmit user data to a third-party MCP service.

Static analysis

No suspicious patterns detected.