ClawHub
Back to skill

Security audit

Book Inspection

Security checks for vulnerabilities and agentic risk

Overview

The skill is a coherent inspection-booking helper, but it can send a user's contact details to an external booking service without explicit consent or clear activation limits.

Review this skill before installing. It appears intended for legitimate inspection booking, but users should be told when their name, email, phone number, location/ZIP code, and selected appointment details will be sent to Lokuli, and the agent should ask for explicit confirmation before creating any 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)

Vague Triggers

Medium
Confidence
95% confidence
Finding
The trigger description is broad enough to activate on generic 'inspection' requests without clearly constraining the domain or requiring user confirmation. That can cause the agent to invoke this booking skill in situations where the user only wanted information, increasing the chance of unintended third-party contact or data flow to the external Lokuli MCP endpoint.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill includes a booking flow that transmits personal data such as customer name, email, and phone number to an external MCP service, but it provides no warning, consent step, or data-handling notice. In an agent setting, this creates a meaningful privacy risk because users may not realize their contact details are being collected and sent to a third party as part of tool execution.

Static analysis

No suspicious patterns detected.